I If WIWITCIII” . I. . .... III_I ”LIL? 'III I 'I"II' i‘ ‘H'mm‘l “I LII III ;.hllrn]~1 'I; f1; 11 5 ._ ..... I:I‘: I :::: ‘‘‘‘‘ . ‘u- ' - , . ~ —. .4 — > . a I . - - - . . A ,. ~ ‘, -. 4 .. . . . IA . .—. .. _ - o u v .. >«r - f; . . . '— . - - J. n - . -7” . 7 -I w- M an: “ I-‘ - L'n. . ¢ —‘ ' .. -_. - £._ 7.. on 4 v u. ‘I ,‘I _ .‘ ‘- _ a .. I . . »‘ A ;-. ’ - -«n- fl " p' ' .— , . . . - .i, I . o ., . . . ‘ ”3 . . ' - . . . . . ’ ‘ ' a ‘ no a ' ‘ -. ’ < 'fi--"'= aw.- -‘-: ‘ . I. ' .. .. ;r v - . .. *— . . . ,. -. ”I. . v .1 > .~ ' -I v - ‘ ¢. A , _ _ .-_- . - ... . - -x: . k...” . .- A m - ~ - .. é! ,J a - ‘ -. . “v -,. . o . .. 2:! _I v...- — . ‘I T: “4-7 1; . ‘0 _ .. _. .~ _ . - . ~_.--..-¢.= : . ’ n , . . A , ‘-‘, :1 .. ’ ’ ‘ V.- '1. —.-I 4‘ . ,v a- - .. .. - _. - '3. m . - . . -- . ~ ~ . - -— ~— ,- . . C - _ V” - —- - , I. I ' n . - . Ilu‘ ' . "I ’2: 4 .,’ ,. '31 . I .a. . ' . . 3-4 a l .J a In r. f , AH - r . fi. - - 1. '- yv‘ ' . u - 'v . ~ . n - ~ -.- » l ‘ . I: _ _ I. :u , ‘ , - V ‘9 , . . o a 1” -4 ." n 3-. _'A 'r 4 (I. 0- - --u I" u n I O . ‘ _ ‘ 7‘ r - “, II I ELI: .HLI IIIt "“I“ ';. I‘ 7'3? 3 ' .I I! I"; ' IifI ' I 3‘ I II III I I ,1 , ’ I I 1,1'5 F ‘I I 1 iII W -I I;.:..II '1‘? ”I IIIIIIy’fi‘J}"I1I i" I II“ :1; I! LII‘IEI: II III I I . 'III I' I' IIIIIIIIIIIII I ,II IIIII IIIIIIIIIIIII. IIIIIIIII I I. "II “ . I .I‘r . I ' I I I I',',I I ; I‘ ' ""1I ' I x .‘ If" "‘II'I‘ . 'I‘ 'III‘ ll‘ .' I'I‘fll'u‘ "23 .31"!I '4' ’I‘I' : III" :~IIII‘ I I I. II :I I: II I I Ii _‘ III ‘I' II, ‘II‘II II II"I§I.II'IZ‘ III 'II I"II'I'!U:;I .II‘HII‘H u I "I . " II I .. I I.,- II, I I EI'I-L‘i II , I III ‘II III III...» 1IIIIaIII I I II'III| II II III‘ ‘III III,' I I‘: II I III".‘.I.II I III}; -: l' V:'3|I:I‘;III{ ‘ItII‘III I ‘II I I‘I I I “I! am" 'I‘I'Iitv "‘.I‘ ”MI-Inn“, ”III , I - I II ': II ' I:"I I' I' III I I‘ {III IIIIII III' I_I' I' I‘I' IIII IIII ‘I‘I'II:I I'I II III’IIIWIII .IIIIH! HI. I I I“ I II “II H.‘WIIIRHQI'EHW‘du‘h“1’44I“if.“I‘IJIIIEI I} 'I ‘ I , II I M )‘I- I :M‘ ".' ‘ " I“ ' II II ’II'II I III“I--‘-£I I 'l’ 5 HI: ." HI, "'1" I II "‘ "1"" | I. I ll 2W Hi" i A will}: ‘t'LI {I‘II‘III‘.‘IIII:‘I: ‘I‘RI’E‘IIifl‘ I‘ I'Il‘:{"3.;|Il.'I!K‘§>..‘l. l «:11 IE‘IIiIEl I ‘1‘.“ ['1III’I‘ Lh'IIHM’ILQHII ‘IIUI’I‘I‘. ’I‘ II INIZ,‘ ‘. .‘ mII‘III "II'IIII l" '1 ’ IIII I I" II" III III ,IIle'I III. I . I-l IIIIIIIIIIII‘ ‘ II III II IIIVIIII IIIIIIIII II‘IIIIIIIIIII I.IIIII1I IIIII ‘III III IIIIII‘II I.“ ‘IIII L5” ”1! I” II! IIIIWIII I II IIIIHII :III I III I'III IIIII‘WI WI hI‘IEIlI ”I?! IIIIIIIII' “‘}1““‘III!‘E~II“ II III I fly I I" M ' t“ .IIIIII III' III MI ""III ___;:: LIA :’.__ W; _____._- v? 2:...» ”— LIBRARY . :1) £3 4 5431.? 3 Michigan State Univ'cnity This is to certify that the thesis entitled Fluidity And Chromophore Interactions In Purple Membrane: Electron Spin Resonance, Fluorescence And Circular Dichroism Study presented by SHARMI LA SHARAD GUPTE has been accepted towards fulfillment of the requirements for i \ t w A 0 degree midi/air! ‘9 2: 2 l 6 gig I 2: 7 Major profess Date ?/l’ (2} Iggy 0-7 639 I i I I 1:; iv ~ Nu? Clark-3‘, W91”: :3 rs” (.inh’fis 1-! - . _ t'HZ,LI I : ., , ‘. ‘ ui'L'."‘Th‘ , I ‘ V r . for fine degfy¢ ;f SMMI LA sum GUPTE DOCI‘UH 0F mu'am ALL RIGHTS RESERVED fitment. of a: my. 1-4-4 um ' _. . ‘_ ' -,- . 3‘, - ’ ‘x: - - .i a . a.‘ 5;: _ ~_ ELUIDITY AND CHROMOPHORE INTERACTIONS IN PURPEE ‘ MEMBRANE: ELECTRON SPIN RESONANCE. FLUORESCENCE AND CIRCULAR DICHROISM STUDY ‘ By 1 Sharmila Sharad Gupte 1 (Sheila Kamalakar Karkhanis) A DISSERTATION Submitted to .tgé Michigan State University h 'in partial fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biophysics 1978 (9 Copyright by Sharinila Sharad Gupta (Shgila Kamalakar Karkhanis) ‘19? ...f,,...; ." , .,.. - . FLUIDITY AND CHROMOPHORE INTERACTIONS IN PURPLE MEMBRANE: ELECTRON SPIN RESONANCE. FLUORESCENCE AND CIRCULAR DICHROISM STUDY By 'Sharmila Shared Gupte (Sheila Kamalakar Karkhanis) AN ABSTRACT OF,A DISSERTATION Submitted to Michigan State University in partial fulfilment of the requirements. for the degree of DOCTOR OF PHILOSOPHY i' Department of Biophysics “ 5-3.: 1 ~ 1978 ABSTRACT FLUIDITY AND CHROMOPHORE INTERACTIONS IN PURPLE MEMBRANE: ELECTRON SPIN RESONANCE, FLUORESCENCE ,AND CIRCULAR DICHROISM STUDY By Sharmila Sharad Gupte (Shaila Kamalakar Karkhanis) -Halobacterium.halobium is a halophilic bacterium which requires 4 M NaCl in its growth medium. It exhibits differ entiated purple membrane (pm) regions in the cytoplasmic membrane. The pm forms a rigid lattice of two-dimensional crystalline array. Bacteriorhodopsin is the only protein in the pm, its chromophore is a retinal covalently bound to an é-amino group of a lysine via a Schiff base. The circu- lar dichroism (CD) spectrum of pm shows a positive and a negative band. indicating exciton interaction between the chromophores of protein trimer. Sharmila Sharad Gupte In this study, we have examined the fluidity of the pm and cell membrane vesicles (cmv) using spin and fluore- scence probe fatty acids. These probes have shown that the omv are rigid, although less rigid than pm. Nitroxyl labelled (5—NS) pm exhibits a phase transition at about 31°C whereas the cmv show a phase transition at about 22°C. In cmv, the rigidity extends at least to the twelfth carbon position reflecting a tight packing. Relative fluidities of the pm and the cmv can be measured also using changes in the fluorescence wavelength maxima of anthroyl stearate (AS) probe. The emission maxima of AS probe in phosphatidyl choline vesicles, cmv and pm occur at 4A6, #37 and 435 nm respectively. The blue shift in the emission maxima reflect an increasing degree of rigidity of the membrane. In the pm. there is an additional peak at #17 nm. This probably arises from AS probe molecules in the pm which are tightly packed such that the anthracene moiety is twisted out of plane with respect to the carboxyl group. There are two populations of AS molecules in the pm as detected by fluorescence decay measurements, one near bacteriorhodopsin and undergoes efficient energy transfer to the chromophore. The other population lies in the bulk lipid phase. The rigidity of the pm seems to be due to a high ratio of proteinzlipid and interdigitation of protein in the membrane. Our studies of chromophore-chromophore exciton inter- action using absorption spectroscopy confirms published Sharmila Sharad Gupte CD data. Stepwise reconstitution of a bleached pm shows that the absorption spectrum of the monomer lies at shorter wavelength compared with the trimer. The orientation of retinal seems to be head to tail at about 20° from the plan plane of the membrane. We interpret the optical activity around 355 nm in bleached pm as that retinal oxime occupies the cavity even after bleaching. Our results show that the retinal oxime can be removed by retinal during the reconsti- tution. We have succeeded in obtaining a fluorescence ana- log of bacteriorhodopsin by incorporating diphenyl hexa- triene (DPH) in bleached pm. The DPH replaces the retinal oxime. Such an approach may prove valuable in exmining the nature of the chromophore cavity. A photophysical study of a fluorescence polarity probe (dansyl sulfanoamide) provides data on its emission proper- ties (quantum yield, emission maximum, fluorescence life- time) in media of different polarities and in ethanol as a function of temperature. This study required us to obtain solvent contraction factors at different temperatures. Our study of the emission properties of anthroyl stearate in different solvents at room temperature and at 77°K show that in addition of being a polarity probe, it is also a fluidity and a "packing" probe. Our study has pointed out numerous new directions for future work on H. halobium and its purple membrane. ' PTNORIL'IZPENTS I would like: 3... ':I..:.efi'd .2.) use. .;,:;:e;iation to or; “viser. Dr. 51. .14.. , v..- 3-4 «fiance and under- ‘iif Wag. iithtit his .rp;.clx‘ and pan-rial af’ertion. this may night 4.: Tee: «cg; ’ «.rgr'fiate the my parents. family -mwcm giwn far t'reand frihndsc: 1.9 :Jsertuion topic. who helped to be ”It 01' this v.21: wa— whateIiam today." '- laboratory. I On ”bind to Dr. Bang 5.: gu: Jan-v _. an; p art and use of fit nu. faciliiies. Alec. I wculd like to thank Dr. Ind understanding during the course of the study. I like to thank my disswts. ion emit see one.“ 0! “ ”Fungi. hug. mGrcarty and Popav for their “CM '0”. I would like to thank Dr. J. r. com. Biophysics dept for his amiss or Wig; Wu for her advic or. various matters and her triad: _.\ l ACKNOWLEDGMENTS I would like to express my deep appreciation to my adviser, Dr. M. A. El-Bayoumi, for his guidance and under— standing. Without his support and parental affection, this study might not have been completed. I appreciate the freedom given for the selection of the dissertation topic. Part of this work was done in Dr. Haug's laboratory. I am indebted to Dr. Haug for his guidance, support and use of the PRL facilities. Also, I would like to thank Dr. McGroarty for her advice on various matters and her friend- ship and understanding during the course of the study. I would like to thank my dissertation committee composed of Drs.El-Bayoumi, Haug, McGroarty and Popov for their useful suggestions. I would like to thank Dr J. I. Johnson. Chairman, Biophysics dept for his advice on departmental matters. I would like to thank my colleagues and friends David Carr, Dr. Chang Chen, Dr. David Johnson, Barbara Kennedy, Denise Mazorow, Gary Smith and Herman Weller for making this study memorable. Special thanks to David Carr for his expert assistance with Time Resolved Spectrofluorimeter v _.__—_ v and sptropolarimeter and to Denise Mazorow for typing the rough draft of the dissertation. I would like to thank Dr. Zand, University of Michigan for the use of the Spectropolarimeter. I wish to express thanks to my husband, Dr. Sharad Gupte for his patience and understanding. Without his efforts, this study could not have been initiated. Also to my brother, Rajiv Karkhanis for his help. Finally, very special thanks to a very special person, my son, Naren for enduring through the course of this work. This work was supported by Department of Human Medicine and Department of Osteopathic medicine, Michigan State University. Chapter 1 Chapter 2 W Introduction .III...IIUIIOIOIICIOICIIOOOI. Section I-Structure and function of biological membranes (a) (b) (C) (d) (e) (f)Mobility of proteins in the membrane .. Introduction ......................... Functions of membranes ............... Membrane asymmetry ................... Nature of protein associations with membranes ....................... Membrane models ...................... Section II—Fluorescence and electron spin resonance probes of membranes (a) (b) (c (d V V Spectroscopic techniques .............. Physical parameters of fluorescence probes ................................ Physical parameters of electron spin resonance probes ...................... Applications of fluorescence and ESR probes for membrane studies ........... vi 13 1A 19 23 Chapter 3 Chapter 4 vii A literature review of the purple membrane (a) Characterization of H. halobium ....... (b) Structure of purple membrane .......... (c) Comparison of bacteriorhodopsin with visual pigment (rhodopsin) ............ (d) Intermediates of bacteriorhodopsin .... (e) Conformation of retinal in bacteriorhodopsin ..................... (f) Fluorescence of the chromophore in bR . (g) Chromophore interaction in bR ......... (h) Light induced conformational changes in bacteriorhodopsin investigated by cross—linking technique ............... (i) Proton gradient across the cell membrane of H. halobium ............... (j) Light induced active transport of amino aCidS 0......IOIIIIIOOOIIOIIOOIOI (k) Biogenesis of the purple membrane ..... (l) Phototaxis in cell containing purple membrane 0......UIOIIIICICDOOIIIIIDOOI. Methods and materials Methods: (a) Culture conditions for Halobacterium halobium OIO...ICOOIIIOOIICIOIUCIOOI0.. 29 32 36 38 #0 Ah #4 “5 #8 49 49 51 Chapter 5 viii (b) Harvesting ............................ (0) Isolation of H. halobium cell membrane vesicles ..................... (d) Isolation of purple membrane .......... (e) Criteria for the purity of the purple membrane ....................... (f) Preparation of liposomes .............. (g) Purple membrane liposomes ............. (h) Bleaching of purple membrane .......... (i) Reconstitution of purple membrane ..... (j) Solubilization of purple membrane ..... (k) Gel chromatography .................... (l) Labelling of membranes ................ V (m Temperature variation studies ......... V (n Spectral measurements ................. Materials cola-conic.UOIIOIOOOIIIoOoeonoIOO Photophysics of a polarity probe (dansyl sulfanoamide) and a packing probe (anthroyl stearate) (a) Introduction .......................... (b) Contraction factor .................... (c) Dansyl sulfanomide .................... (d) ArIthroyl stearate IOCOIOICOOIOCOICIIOIQ- 54 54 55 56 59 6o 60 61 62 62 65 65 66 72 7h 75 8o 86 ix Chapter 6 Fluidity studies in purple membrane and Chapter 7 cell membrane vesicles using electron spin resonance (ESR) and fluorescence probes (a) Introduction .......................... 95 (b) Fluidity and phase transition tempera- tures of purple membrane and cell membrane vesicles using spin probes ... 95 (c) Fluidity and packing of purple membrane and cell membrane vesicles using a fluorescence probe, anthroyl stearate . 103 (d) Review of the structure and composi- tion differences in lipids and proteins of purple membrane versus cell membrane vesicles ...................... 111 (e)SumaryIIDIICIIICIIIIOIIII.IOII'IIIIII113 Chromophore interactions of bacterio- rhodopsin of the purple membrane (a) Background ............................ 115 (b) Some theoretical considerations for the absorption of retinal-Schiff base . 116 (c V Review of the optical activity of bacteriorhodopsin...................... 118 (d V Theory of exciton interaction ......... 119 Chapter 8 x (e) Experimental absorption and circular dichroism data of bleached and step— wise reconstituted purple membrane .... (f) Fluorescence of retinal oxime ......... (g) Absorption, fluorescence and circular dichroism of fluorescence analog of bacteriorhodopsin...................... (h) Summary OI..-IIOIOIOIIOODOIIOIOUIOI.... Future work (a) Effect of environment on the growth and membrane structure/composition .... (b) Nature of the retinal binding cavity .. (c) Conformational changes in bR due to a photon absoprion .................... (d) Protein and protein-lipid interactions .......................... (e) Relation between exciton interaction and function of bacteriorhodopsin ..... Bibliography 00..........IIIOCI....II..OIII 121 132 134 139 142 143 146 147 148 149 LIST OF FIGURES Figure 2.1 Spectral parameters of electron spin resonance ...-IOOIIIIIIOOIIIIOCOIIOOOOII 20 Figure 2.2 SpeCtral parameters of electron spin resonsnce ...-IIICOIOCICIIIOI...-....IOUI 2“ Figure 3.1 X-ray diffraction data for the structure of purple membrane ........... 35 Figure 3.2a Chromophore of bacteriorhodopsin ....... 41 Figure 3.2b Model of bacteriorhodopsin intermediates 41 Figure 4.1 Growth curve of H. halobium ............. 53 Figure 4.2 Absorption spectrum of the purple membrane from bacteria grown in our laboratory ............................. 58 Figure 4.3 Absorption spectra of bacteriorhodopsin under various conditions of bleaching and reconstitution ..................... 64 Figure 4.4 Schematic of temperature control accessary for the temperature between 100°K and 3oo°K ........................ 67 xi Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure F5 ure Figure 4.5 5.1 5.2 5.3 5.4 5-5 5.6 5-7 5.8 6.1 6.2 6.3 6.4 xii Nanosecond time-resolved spectr Spectrofluorimeter ..................... 71 Contraction factor ..................... 78 Refractive index of various solvents ... 79 '3' factor for various solvents . . . . . . . . . 81 Emission maxima of DNSA in various solvents ............................... 83 Emission properties of DNSA in various solvents .................... 8? Emission spectra of AS in 3-methyl pentane (1) room temperature and 0 _ 77 K. ex-r365nm............o......... 91 Emission spectra of AS in ethanol at (1) room temperature and (ii) 77°K ..... 92 Emission spectra of ACA in ethanol at (i) room temperature and (ii) 77°K ..... 94 Structure of electron spin resonance and fluorescence probes for the fluidity studies ....................... 96 A typical ESR spectrum of 5-NS in pm ... 9? Plot of 2T// of 5-NS in (1) pm and (ii) cmv as a function of temperature .. 99 Plot of 2T// of 12-NS in cmv as a function of temperature ................ 102 \I. Die Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure 6-5 6.6 6.7 7.2 7-3 7.4 7-5 xiii Emission spectra of 10-5 M AS in (a) pcv in liposome buffer (b) cmv in salts 2, and (0) pm in distilled water ..104 Decay of AS fluorescence in various membranes .............................. 106 The temperature dependence of the emission maxima of AS in (i) pm and (ii) cmv cocoon-coocute-cuouoooooccuooue 109 Absorption spectra of stepwise bleaching of pm ........................ 124 Absorption spectra of stepwise reconstie tution of bR with all-trans retinal .... 12? Circular dichroism spectra of step- wise reconstitition of pm with retinal . 131 Fluorescence spectra of retinal in (i) bleached pm (ii) reconstituted pm and (iii) native pm ................. 133 Structure of some polyenes which may occupy the retinal cavity .............. 135 Absorption spectra of DPH—bR analog and control bleached pm with equimolar retinal ................................ 13? Circular dichroism spectra of DPH-bR and control pm ........................ 138 Fluorescence spectra of DPH-bR in (i) bleached pm and (ii) reconstituted pm .. 140 .él‘c 3‘,“ If; gable. ..5- 1 . LIST OF TABLES .Emission properties of DNSA in. . . t , 1.- various solvents ....................... 84 ~Emission properties of AS in various solvents at room temperature. “max: 365 m. assess-oncogenes...net-ere. , 89 CHAPTER 1 INTRODUCTION A gram negative bacterium, Halobacterium halobium grows optimally in near saturating salt concentrations. In the presence of light and at low rates of aeration, the bacterium forms distinct purple patches contiguous with the cytoplasmic membrane. When the NaCl concentration of the suspending medium is lowered to 1.0 M NaCl, the cell wall, which is made of two layers of protein, disintegrates and closed cytoplasmic membrane (cmv) remain intact. Dialysis against deionized, distilled water yields purple membrane (pm) patches, rest of cytoplasmic membrane solubilizes (Oesterhelt & Stoeckenius, 1971),(Stoeckenius & Rowen, 1967),(Blaurock et al, 1976).The difference in the stabi- lity of pm and cmv to water was investigated in this study by using spin and fluorescence probes. The pm is a very interesting biological membrane to study. It has only one protein bacteriorhodopsin (bR) which makes it uniform in structure compared with other biological membranes. Therefore, X-ray diffraction and electron microscopy of the architecture of pm have yielded valuable information about this membrane which may be extended to other membranes. These techniques show that the pm is composed of.a rigid lattice of two dimensional crystalline array of P3 symmetry. The crystalline lattice of pm indicates rigid 1 2 membrane which is substantiated by our ESR and fluorescence study of pm using fatty acid probes. Electron microscopy of tilted unstained pm shows that 70%-80% of the protein of the purple membrane, bR, consists of 7 x-helices, almost perpendicular to the plane of the membrane. Three molecules pf bR form a trimeric unit which may be responsible for the hexagonal lattice structure of pm seen with electron microscopy (Henderson & Unwin, 1975). Recent studies with freeze-fracture of pm show that the interaction between protein entities may extend to between 9 or 12 protein molecules (Fisher & Stoeckenius, 1977). Our studies of the pm in solution also indicate that the pm is a very rigid membrane and most of the lipids seem to be boundary lipids. The chromophore of bR, which absorbs at 560 nm, is a retinal covalently bound to an e—amino group of a lysine, forming a protonated Schiff base with the protein (Oesterhelt & Stoeckenius, 1971). In native conformation, the retinal can occur as a 13-cis or an all-trans isomer. Absorption of a photon causes some conformational change in the protein, the Schiff base becomes unprotonated and a vectorial translocation of the proton takes place. This intermediate absorbs at 412 nm, and is called "bleached" pm as its purple color disappears. Within milliseconds, the pm converts to its native conformation, therefore, at moderate intensities, the 412 nm complex is undetectable. (Oesterhelt & Hess, 1973).Irradiating the pm with high intensity light between 500 nm and 700 nm and in the 'Y‘J (J s." a a 9'. a I 3 presence of 0.2 M NHZOH, pH 7.0 causes breaking of the covalent bond of retinal from the protein. The complex can be reconstituted by the addition of retinal in the absence of NHZOH. Retinal is thought to be in a protein cavity which is unaccessible to solvent without light. Light induced conformational change makes it accessible (Konishi & Packer,1976). Our results indicate that retinal oxime occupies the "cavity" even after bleaching, however, it iS displaced by retinal during reconstitution. We have succeeded in substituting retinal by a linear polyene, diphenyl hexatriene (DPH) which can be used as a fluores— cence analog of bR. Another important aspect of pm is the chromophore- chromophore exciton interaction between adjacent retinals (Bauer et al, 1976). Our results from absorption studies of stepwise reconstitution agree with the circular dichroism data. This interaction may be important in increasing the efficiency of photon absorption. The absorption of a photon causes a vectorial translocation of protons across the membrane, resulting in a proton gradient making the medium acidic. The light induced proton gradient is used by the cells for ATP synthesis. The author is intrigued by various aspects of the H. halobium membrane, particularly pm. The complexity of the biological membranes and various functions performed by them are, in author's opinion, most important and contain interesting facts in terms of understanding a "living" 4 cell. Especially the pm, the degree of sophistication of its structure and function compared even with the visual pigments makes her wonder whether H. halobium (as well as other halobacteria species with a purple membrane) has evolved further compared with the rod outer segments. Different probe molecules have been used to understand the structure and function of various biological macro- molecules.and membrane systems. An extensive study of the photophysics of fluorescence probes under various conditions is essential in order to extract meaningful information from the fluorescence studies of biological membranes. According to the above rationale, the research report- ed here is divided into three parts: (A) A photophysical study of (i) a polarity fluorescence probe, dansyl sulfanoamide (DNSA) and (ii) a packing fluorescence probe, 12-(9anthroyl) stearate (AS) (Chapter 5). (B) The relative fluidity studies of pm and cmv using ESR and fluorescence probes (Chapter 6). (C) The investigation of the chromo- phore binding site and chromophore—chromophore interactions using fluorescence, absorbance and circular dichroism (CD) techniques (Chapter 7). Future directions for the conti- nuation of these studies are outlined in Chapter 8. In addition, a chapter on the membrane structure- function and a brief description of the techniques used in this study (Chapter 2) is included. A comprehensive litera- ture review seems a necessity and, therefore, is included as Chapter 3. A: CHAPTER 2 SECTION I STRUCTURE AND FUNCTIONS OF BIOLOGICAL MEMBRANES (a) Introduction: Biological membranes play a crucial role in almost all cellular phenomena, yet, the understanding of the molecular organization of the membranes is still rudimentary. Once the plasma membranes were thought merely to be a barrier between the cytoplasm and the outside environment of the cell. However, now it is widely known that the plasma membrane is one of the most essential component of the cell. (b) Functions of membranes: Some functions attributed to the procaryote plasma membrane are (1) active transport (ii) electron trnasport and oxidative phosphorylation (iii) secretion of exocellu- lar proteins, toxins and enzymes (iv) protein synthesis and membrane associated ribosomes (v) phospholipid and glycolipid biosynthesis (vi) DNA anchoring, replication and cell division (Salton, 1971). In addition to these functions. specialized membranes of eucaryote cell parti- cipate in functions such as cell-differentiation, cell-cell rEcognition, contact inhibition, density inhibition, synaptic transmission, drug and/er hormone interaction etc. ftij Widelfian 1976). The diversity of the functions of these ’5, ibmbranes reflects the structural differences, 1. e. the zifipfiiation of the components of the membranes. For example, :v:?t;§{ 6 the erythrocyte plasma membrane, which is a well-studied fluid mosaic membrane, contains 40% lipids, 52% proteins and 8% carbohydrates by weight (Steck, 1974). On the other hand, highly ordered structures of mitochondria and chloro- plasts contain 80% proteins and 20% lipids. These membranes are responsible for the energy transduction mechanism. Similarly, the purple membrane contains 75% protein and 25% lipids (Oesterhelt & Stoeckenius, 1971) and is a highly ordered two dimensional crystalline structure which probably correlates to its photocoupling (Oesterhelt & 4‘ Stoeckenius, 1973) and photosensing (Hildebrand & Dencher, ii 1975) function. From these examples, it seems that the higher the percentage of proteins in the cell membrane, the more specialized its function becomes. The matrices of these specific membranes must be highly organized to perform certain functions. In other words, there is likely to be a relationship between the high content of proteins and specialized membrane function. (e) Membrane asymmetry: Another aspect of structure-function correlation of biological membranes which has been emerging in recent years is the asymmetry in the membrane lipids (Gordesky, 1976) and proteins which may in turn be related to its function (Rothman & Lenard, 1977),(Wisnieski & Iwata, 1977). Although the membrane asymmetry has been discussed for many years, it is only recently that direct evidence has been “” obtained confirming the asymmetry at the molecular level. 7 Chemical modifications of exposed residues of the membrane proteins in the intact cells by some impermeable molecules shows that only the external surface of the cell is labelled. However, similar chemical modifications of leaky cells (permeable) show labelling of some other proteins in addition to the proteins in the intact cells. Also, the chemical modification of right-side-out and inside-out vesicles shows that different proteins are labelled in the two cases. These experiments indicate that some proteins on the external surface of the membrane are diffirent than the ones on the internal surface. In addition, the histo- chemical labels which can be visualized in an electron microscope and are specific for a particular enzyme can be used to investigate the sidedness of its active site. Ferritin labelled antibodies can be used to determine the location and orientation of antigenic determinants in the membrane(Singer, 1974). These studies have shown that the asymmetry of the membrane proteins (i) may be an important factor for Peter Mitchell's chemiosmotic hypothesis of vectorial ion transport across the membrane (ii) may be utilized in transport of.metabolites across the membrane. (iii) may play an important role in membrane biogenesis (iv) may have different control mechanisms for a normal and a transformed cell (Nicolson, 1976). The glycoproteins and glycolipids are detected exclusively on the outer mono- layer of the plamsa membrane where they serve as a link between the cell and its environment. In addition to 8 protein and carbohydrate asymmetry, the phospholipid asymmetry is also becoming evident in lipid bilayers and in natural membranes (Gordesky, 1976). The phospholipid asymmetry may be due to the difference in the charge of the polar group, however, its function in the biological membranes is still obscure. (d) Nature of protein association with membranes: In general, the proteins associated with a membrane can be classified into two broad categories. (i) Peri- pheral or extrinsic : The extrinsic proteins are mainly bound to the membrane by the 'ionic' forces, therefore, they can be isolated by changes in the ionic strength, pH, etc. (ii) Integral or intrinsic proteins: These integral proteins are essentially lipoproteins in which the lipids are closely associated with the protein and are essential for its function or conformation (Capaldi, 1974), (Singer & Nicolson, 1972),(Farias et al, 1975). Depending on the asymmetry of the membrane, the intrinsic proteins may be further classified as (1) ectoproteins which have their hydrophilic mass projecting beyond the extracyto- plasmic surface (outside of plamsa membrane) of the lipid bilayer, (2) endoproteins which have most of their mass associated with the cytoplasmic side of the membrane (Rothman & Lenard, 1977) and (3) transmembrane proteins which asymmetrically span the bilayer from the intra- cellular to extracellular space (Henderson & Unwin, 1975), (Singer, 1974). The function of these transmembrane 9 proteins seems to be mainly the transport of ions and metabolites across the membrane by either conformational change or by a 'pore' mechanism. (e) Membrane models: From the above discussion, one may get a glimpse of the diversity of the membrane structure. Membrane models which may account for most of the experimental observations are discussed briefly. The early membrane models consisted of a 'unit membrane' hypothesis in which all the lipids were in a thermodynamically favorable bilayer and the proteins were composed mainly of @-sheets which may extend in the bilayer to form a 'pore'. Although birefringence and small angle X-ray diffraction studies supported the notion of the unit membrane and stained electron micrographs of various specimens showed characteristic double tracks of unit membrane, later studies have questioned the validity of the methods used (Wallach & Winzler, 1974). The interaction of the intrinsic proteins with surrounding lipids is mainly hydrophobic contrary to the ionic interaction of the fi-sheets. Singer-Nicolson's fluid mosaic model (Singer & Nicol— son, 1972) hypothesizes a fluid lipid bilayer inter- digitated by proteins. On thermodynamic grounds, the non- covalent hydrophilic interaction among the polar heads of the phospholipids and the polar residues of the proteins and the hydrophobic interaction between the fatty acid 10 chains of lipids and the non-polar residues of the proteins, seem favorable. Also, lipids, (mainly their hydrophobic interaction) are essential for the activation of some, if not all, of the intrinsic enzymes which seem to stabilize the active conformation of the enzyme (Overath et al, 1976). According to the fluid mosaic model, most of the lipids and proteins are capable of rotational and translational diffusion in the membrane matrix (Lee, 1975). A major contribution to this model has been made by Frye and Edidin (1970) who investigated the membrane receptor properties of a cell—fusion heterokaryon. Individual cell antigens were labelled with different antibodies and allow- ed to diffuse after fusion. After 40 minutes, intermixing of two labels occured indicating fast diffusion of two types of antigen at growth temperature due to a fluid, mosaic membrane. Even in the fluid membrane, as much as 30% of the lipids may be immobolized (boundary lipids) due to a strong hydrophobic interaction with the proteins (Stier & Sackmann, 1973). Although some membranes seem to fit in the fluid mosaic membrane model (e. g. rod outer segment membrane), others seem to be highly ordered membranes in a rigid crystalline lattice. The electron micrographs of a gap junction between the cells of mouse liver and the purple membrane of Halobacterium halobium both show a two-dimen- sional, rigid, hexagonal crystalline lattice (Goodenough & Stoeckenius, 1972),(Blaurock & Stoeckenius, 1971). The ”a 11 boundary lipids of these membranes may comprise most or all of the lipids restricting the diffusion of both proteins and lipids. The polarization studies of the protein mobility in the purple membrane have shown that the protein appears to be particularly immobilized and has a rotational relaxation time of at least 20 msec (Naqvi et al, 1973). This is contrasted with the rotational relaxa- tion time of rhodopsin of about 20 psec (Cone, 1972). (f) Mgbilityiof proteins in the membrane: In addition to the immobilization of proteins and lipids due to ordered structure, other mechanisms also alter the mobility of the membrane components. These are (i) Supramolecular functional aggregates: In case of erythrocyte ghosts, spetrin forms supramolecular aggre- gates which move as an aggregate due to cross-linking by spectrin's antibody. Also, in the mitochondrial inner membrane, a tight association between five different proteins makes uniform arrays throughout the membrane. This supramolecular association may increase the efficiency of electron transfer and its coupling to ATP synthesis (Capaldi, 1974). (ii) Ligand-induced rearrangement of proteins: 'capping' of cell surface receptors by lectins or antibodies may change the diffusion properties of the receptor proteins (Bretscher, 1976). (iii) The glyco- proteins and glycolipids may have carbohydrate crosslinking and extensive ionic and hydrophobic interactions at the outer monolayer of the asymmetric membrane. These 12 I interactions may change the diffusion of the membrane components significantly. If this is the case, the diffu- sion of an endoprotein may be higher than that of an ectoprotein and the diffusion of a transmembrane proteins of similar dimensions may represent an average diffusion. (iv) Divalent cations interact with lipids and proteins which causes crosslinking and structure fomring (Hauser et al, 1976). These effects decrease the mobility of the membranes such as rat synaptic plasma membrane (Breton et a1, 1977). (v) Finally, the cytoskeletal attachment of microfilaments and microtubules may restrict and regulate the movement of proteins in the membrane (Linden & Fox, 1975). The cytoskeletal attachment exhibits a transmembrane control of the cell-surface receptors and may be very important for the surface modulation in cell-recognition and in cell-growth (Edelman, 1976). ‘r———_——"_ SECTION II FLUORESCENCE AND ELECTRON SPIN RESONANCE PROBES OF MEMBRANE (a) Spgptroscopic tgghniqugg: It is evident from the discussion in the preceding section that the biological membranes are very complex and that various techniques need to be combined to investigate the structure and the function of the membranes. These techniques range from electron microscopy to the chemical analysis of the lipids. Spectroscopic techniques have been used widely to gain information about the membranes. X-ray diffraction and recently, electron diffraction of tilted, unstained specimens have been most successful for structure determination. However, the information obtained by these techniques is mainly static and time-averaged. The mecha- nism by which the structure enables the function cannot be determined by these techniques. Spectroscopic probe techniques provide insight into the correlation of the biological structure with its function. The techniques involved in the membrane structure-function relationship are (i) absorption (ii) emission (iii) optical rotation (iv) Raman and resonance Raman (v) nuclear magnetic resonance (NMR) (vi) electron spin resonance (ESR) and (vii) various aspects of the magnetic resonance techniques 6. g. ENDOR, FDMCD (fluorescence detected magnetic (or .m natural) circular dichroism) etc.. The time domains 13 I'-" 14 investigated by these techniques range from picosecond to infinitely long time on the molecular scale. Out of all the techniques mentioned here, the fluorescence and the electron spin resonance techniques are very sensitive. Also, the number of parameters that can be used to obtain meaningful information is very large. Probes of various specificities can be used with these techniques. The term 'probe' implies the penetration of a region by the molecule with no or minimal disturbance of that region. The spectroscopic probes can be divided into (i) intrinsic and (ii) extrinsic 4 categories. Intrinsic probes are the naturally occuring chromophores, e. g. tryptophan retinal, chlorophyll, heme proteins (paramagnetic ferric ion), dopamine IS-hydroxylase (paramagnetic copper). Extrinsic probes are the molecules which are sensitive to its local environemt. These are inserted in a system and the local perturbation is measured by the suitable spectroscopic technique. Some of the physical parameters that can be measured with the fluorescence and the ESR techniques is discussed in the next two parts. (b) Physical parameters of fluorescence probes: (i)Emission maxima: If the dipole moment of a fluores- cent molecule in the excited state is different than that in the ground state, the solvent molecules can relax around the excited state molecule prior to the emission. This results in the emission maximum shifting to a different energy. This type of molecule e. g. 2,6 anilino- 15 naphthalene sulfonate (ANS) shifts its emission maximum from 465 nm for a non-polar solvent to 515 nm for apolar solvent like water. (ii) Quantum yield: Change in the quantum yield from non-polar to polar solvent is parallel to the emission maximum. (iii) Life-timgiof the excited state: Total light emitted after a delta pulse of light is stopped depends on the rates of transitions which depopulate the lowest excited singlet state. For the probes in a homogeneous environment, the decay of the intensity is given by _ ~t/r F(t) =Foe essences-one...noose-s20]- If the fraction of molecules have different degrees of the solvent relaxation, the decay of the intensity is given by n F(t) -t/. —————— =FZ aie 'r‘aoanggeooeeeseeIOOCZOZ F i=1 where ai is the fraction of the molecules which have the life-time of"€i. A semi-logarithmic plot of the first , equation is a straight line whereas for the second equation, it represents a superposition of each straight line compo- nents making it non—linear. The individual life-times can be resolved by deconvoluting the curve by various methods currently available (Ware, 1973),(Kennedy & El-Bayoumi, in preparation). 16 (iv) Polarization: If a fluorescent probe is excited by polarized light, its emission is maximally polarized if. during the life-time of the excited state, it will not have to rotate or change the position. For example, in a very viscous solution or if the probe is rigidly bound by hydro- phobic and/or inoic interaction. The anisotropy is given by A "' ------------ eco-oeso-oee-e-eee-oeeZOB In the case of macromolecules which are rigid sphere and th the probe is firmly attached to it, the decay of aniso- tropy is given by A(t) = Ace-t/¢ no...ooeI-ooosoollOlOeIIZIu where 4) = rotation correlation time, V ¢(seC)=-fl ----- eeeeaeeeuenoee-oeeeelee0e0205 K T V = hydrated volume of sphere, 11 = viscosity, M ( v + h ) v = ------------- .OIIIOOIIIIOIIIOIIOIOIIOOIOIZI6 N a molecular weight, v = specific volume, h ’= hydration, N = Avogadro's number i};n the case of rigid ellipsoidal molecules, “3:523:75 l ' - 1? “522-- .2 fie-(Wm. ..............27 where 'rli 's are related to the rotational diffusion about major and minor axes of the ellipoid. (v) Quenching: Two types of fluorescence quenching can be observed. (a) Static quenching which is an interaction between the probe and other molecule/molecules in the ground state to form non-fluorescent complexes. ——————— = 1 + KS [Q] ..eeeeo-eeoeeoeeeeeeeZeB Where F0 and F are the intensities in the absence and in the presence of the quencher and [Q] is the concentration of the quencher. This type of quenching reduces the intone sity without changing emission properties. (b) Dynamic quenching: This process competes with the emission for the depopulation of the excited state, thereby, decreasing the intensity and the life-time of the excited state. In the case of the dynamic quenching, ‘ F ,—;a--- = 1 + KD [Q] .....................2.9 KD=k‘t', k=4fiaDN' 21o 18 a = sum of the molecular radii N' = number of molecules/millimole D = sum of diffusion constants When both processes are present, 0 --§—-- = ( 1 + KB [0] )( 1 + Ks [p] ) .....2.11 (vi) Energy transfer by Forster Mechanism: The electro- nic excitation energy can be transfered from a singlet excited molecule (donor) to an unexcited molecule (acceptor) if the fluorescence of the donor overlaps with the absorp- tion of the acceptor. This type of energy transfer occurs by a transition dipole-dipole interaction. The rate of the energy transfer is given by 6 -1 uke R- sec eoeooos2012 kDA = 8.7 x 1023 J K2 n' Where J is the overlap integral )FD (Ne, (A) CA)“ .17: J = 01.00.000.02013 .[Fb (70 d) A... K = orientation factor, is 2/3 for random distribution of the donor and the acceptor. r:;,_ .' n = refractive index of the medium 895;. ' 'ks s Irate of donor emission. kDA= 1/QD-1/‘zn. ...-essenceelse-eee00000201u 19 ke = (hp-7713' 2.15 D' = emission in the absence of the acceptor. The distance R between the donor and the acceptor can be calculated by the above mechanism. Other parameters that can be used to obtain meaningful information are (vii) excimer formation (viii) exciplex formation etc.. (o) Physical parameters of the electron spin resonance probes: General discussion: The resonance condition of an unpaired electron in the magnetic field is describes by AE = 11)): gfivp H nonoso...ee-oeo-eeaee-eeeeee2016 With an appropriate condition of the external magnetic field and the corresponding resonance frequency, the energy is absorbed by a paramagnetic sample (Keith et al, 1973). (Fig. 2.1a). For the nitroxide spin probes, the 14N I nucleus has a spin value of 1 therefore, can take the nuclear spin quantum value of 1, 0, & -1. Consequently, the unpaired electron signal is split into three equal compo- nents called low-, mid-, and high-field lines. The absorption of energy is shown in Fig. 2.1b. The spectral parameters are normally measured from the first derivative spectrum, g denotes the center of the spectrum and a is ,if,irthg,hyperfine coupling (Fig. 2.1c). .. I": ._. $— ROW I . ‘ ‘ 4 1"}: ,ft 20 H -) AE =hV= gfipn Figure 2.1a Resonsnce condition for electron spin resonance 1, /\ A /\_ 'Figure 2.1b Absorption of energy by nitroxyl radical go ‘ ‘..?- ificure 2.1c First derivative of absorption of nitroxyl radical, go is mid- field splitting and a0 is width of separation of signal. A? 2.1 Spectral parameters of electron spin resonance 7:154;- 21 (1) Maximum hyperfine splitting: The orbital geometry of the unpaired electron of the nitroxide spin labels is highly anisotropic. The hyperfine structure tensor of nitroxide label can be described by three components of a and a . Components axx and ayy are in the xx' ayy’ zz. direction of the nitroxyl ring and are small compared with azz' The component azz is in the direction of the fatty acid chain by convention. The maximum hyperfine splitting azz is the measure of the restriction of the motion (Fig. 2.2a). (ii) Order parameter: This is another parameter that measures the restriction of the anisotropic motion of the fatty acid spin probes around their long (molecular) axis. The order parameter S measures the deviation of the observed spectra from the case of a complete orientation of the probe. For a completely oriented sample, S = 1, for a random sample, S = 0. a'// — aL T S: CIICIOIIICIl02017 I azz - 0.5( axx + ayy ) T Where T=1/3 (axx+a +a )OOIIIIIOCOIIIIIOOIIOZIIB FY 22 '=‘1 ' +3.. OIIOOCIOOIOIIOOIIOOIIIIIZCi and T /3 (a // J. ) 9 The parameters axx’ ayy’ and azz are obtained from crystal m (rigid lattice )measurements. a'// and a; correspond to 22 the separation of the outer and inner hyperfine maxima (Fig. 2.2b). One can write the equation 2.17 as (Esser & Lanyi. 1973) . ' a - a // _L 8:00.568 ------------ 0.0.0....0.0.0.000000000002020 T! (iii) Rotational correlation time: ESR spectrum of a spin label is extremely sensitive to the rate and the nature of motions the label undergoes. The rate of the 10 seconds for small molecular rotation can range from 10' molecules in an isotropic, fluid medium to 10'7 seconds for fatty acid spin probe in an anisotropic, rigid lattice. For the nitroxyl spin moiety, the rotational correlation time 150 is given by (Esser & Lanyi, 1973) (Fig. 2.2c): h 1/2 10 ---9-> . " 1 00002021 '8 = 6.5 x 10- W0 h 0 (iv) Partition coefficient: The go and the aO tensors in the spin Hamilton of a rapidly tumbling nitroxide A A 5 A goH. 82 + aOI o S 00000000000002.22 We show a small dependence on the polarity of the solvent. Only the high field line is resolved at the 9.5 GHz frequency. Nitroxides like TEMPO with appropriate solubi- lity prOperties partition between the aqueous and the lipid regions of the wet, biological samples. The ratio of the 23 intensities A and B in the Fig. 2.2d is proportional to the relative amounts of nitroxides in the two environments. (v) Prqgimity effects: These can be measured by two parameters (a) spin—spin interaction: Theoretical estimates show that when two nitrogen atoms containing two unpaired 2p electrons approach to within 15 R of each other, the rate of the interchange of their electrons exceeds 107 per second. This rate is rapid enough to cause very pronounced changes in the ESR hyperfine splitting pattern. (b) reduc- tion of the spin label by suitable agents: Suseptibility of the spin probes to the reducing agents depends on the accessibility of the probe. Normally, the fatty acid Spin probe in the hydrocarbon medium is non-accessible to the reducing agents, however, if the probe is in the aqueous environment or if the membrane is leaky, the spin probe can get reduced. (d) Applications of fluorescence angelectron spin resonance probes for membrane studies: The physical parameters for the fluorescent and ESR probes described in the previous section can be used to determine (i) membrane polarity (ii) membrane fluidity (iii) accessibility (iv) distances (v) electrical potentials. (i) Membrane polarity: The most common fluorescent probe used to investigate the membrane polarity is ANS. As discussed earlier, the emission maximumaand quantum yield of fluorescence are the parameters used to determine 24 ll 2a,, A. Figure 2.2a AnisotrOpy of hyperfine splitting Figure 2.2b 1:--- p L 1 Figure 2.20 “5 Figure 2.2d Figure 2.2 Spectral parameters of electron spin resonance 25 the polarity and the change in the polarity for a variety of membranes, e. g. erythrocytes, mitochondrial membranes, bacterial inner and outer membranes, excitable membranes etc. (Azzi, 1975). The location of ANS is shown to be below the polar heads of the lipids. However, ANS has no anchorage in the membrane, making the interpretation ambiguous. Most widely used ESR probes to investigate the polarity are TEMPO or TEMPENE which measure the partition coefficient for the membrane. (ii) Fluidity: Fluidity of the lipids or the biological membrane is a measure of the extent of the packing of the neighbouring molecules which make the membrane matrix. Therefore, it represents the physical state of the lipids. Lipids, especially phospholipids exhibit interesting behavious in the presence of water. The order -£> fluid or gel + liquid crystalline -%> liquid cryastalline transition of the hydrated, lamellar phospholipids can be described as follows: In gel state below critical temperature, T<< T1’ the hydrocarbon chains are in the all-trans conformation, perpendicular to the plane of the bilayer. On increasing the temperature, an endothermic transition occurs which is accompanied by a lateral expansion and a decrease in the thickness of the bilayer. In the liquid crystalline state, T> Th, the hydrocarbon chains maintain an average orienta- tion perpendicular to the plane of the bilayer but are disordered by a rapid trans/gauche rotational isomerization along the chains. Also, there is a simultaneous 26 dissociation of the ionic lattice of the phospholipids by the penetration of water. The temperature of the phase transition in the pure phospholipids-water system depends on the chain length and on the extent of unsaturation. Although the principle is similar, gel-liquid crystalline phase transitions in the lipids of the biological membranes are more complex than the hydrated phospholipids due to (a) the heterogeneity of the chain length and the degree of unsaturation of the fatty acids (b) different polar groups comprising the membrane (c) the presence of intrinsic membrane proteins (d) presence of cholesterol, lipopolysaccharides etc. (e) asymmetry of the membrane. To complicate the matter further, different lipids in the lipid mixtures 'melt' at different temperatures causing phase separation rather than phase transitions. In such a case, the ordered and the fluid lipids form seggregated domains. Spin probes are most widely used to investigate the fluidity of the membranes. The physical parameters used are: the maximum hyperfine splitting 2T/7, the rotational correlation time 'gc and the order parameter S. Most widely used ESR probes are nitroxide labelled fatty acids, phospho- lipids, cholestane etc.. The parameters determined for a fluorecence probe are the emission maximum, polarization of fluorescence, time resolved emission anisotrdpy, excimer emission and fluorescence quenching. The fluorescent probes used to investigate the lipid matrix of the membranes are 26 dissociation of the ionic lattice of the phospholipids by the penetration of water. The temperature of the phase transition in the pure phospholipidsewater system depends on the chain length and on the extent of unsaturation. Although the principle is similar, gel-liquid crystalline phase transitions in the lipids of the biological membranes are more complex than the hydrated phospholipids due to (a) the heterogeneity of the chain length and the degree of unsaturation of the fatty acids (b) different polar groups comprising the membrane (0) the presence of intrinsic membrane proteins (d) presence of cholesterol, lipopolysaccharides etc. (e) asymmetry of the membrane. To complicate the matter further, different lipids in the lipid mixtures 'melt' at different temperatures causing phase separation rather than phase transitions. In such a case, the ordered and the fluid lipids form seggregated domains. Spin probes are most widely used to investigate the fluidity of the membranes. The physical parameters used are: the maximum hyperfine splitting 2T/7, the rotational correlation time vac and the order parameter S. Most widely used ESR probes are nitroxide labelled fatty acids, phospho- lipids, cholestane etc.. The parameters determined for a fluorecence probe are the emission maximum, polarization of fluorescence, time resolved emission anisotropy, excimer emission and fluorescence quenching. The fluorescent probes used to investigate the lipid matrix of the membranes are 27 AS, ONS, DPE, NPN, ANS, pyrene, perylene DPH etc.. Various probes and techniques described so far have shown that the membrane fluidity is the single most important parameter used to investigate membrane structure, contribution of the different components of the membrane, effect of adding or removing ions, ligands, state of energization of the membrane etc.. The resulting changes in the fluidity due to these factors are interpreted accordingly. (iii) Accessipility: The probes bound to the membrane can be used as a possible measure of the accessibility of their binding sites by a number of molecules. Reduction of an ESR signal of a spin probe labelled at the different carbons along the fatty acid carbon backbone has been used to determine the 'leakiness' of the membrane by various agents. The extent of the fluorescence quenching of pyrene by oxygen is used as a measure of the penetration of the lipid matrix by polar molecules. A spin label analog of a local anesthetic was able to quench ANS fluorescence in the erythrocyte membrane indicating a close proximity of the fluorescence probe and the local anesthetic binding site. (iv) Distances: The energy transfer between two fluorescent probes and the saturation effect or the use of biradical spin probes are the two approaches used to determine the distances in a membrane. The aromatic amino acid tryptophan of the membrane proteins is an energy donor for avariety of acceptor probes. The proximity between the protein moiety of the membrane and the DNS-choline in the 28 cholinergic receptor protein purified from the electric organ of Torpedo Marmorate has been established by the energy transfer. (v) Lipid-protein interaction: This is another approach to investigate the membrane structure. The effect of detergents, bleaching etc. on rhodopsin has been studied by using meleimide spin probe covalently attached to the protein. Immobilization of the lipids, especially boundary lipids by the perturbation is also investigated using fatty acid spin probes. CHAPTER 3 A LITERATURE REVIEW OF THE PURPLE MEMBRANE (a) Characterization of H. halobium: An unusual purple membrane is a specialized part of the cell membrane of some extremely halophilic bacteria e. g. Halobacterium halobium (Bergy, 1974), (Larson, 1967). These bacteria need at least 3 M NaCl to be viable, and grow optimally at higher salt concentrations, even in salt crystals. Halobacteria are gram-negative. In normal growth conditions. these bacteria are rod shaped, about one micron in diameter and about 4 - 10 microns long, and have a bundle of flagella at each end. The wild type cells have vacuole membranes which are made up only of proteins. The vacuole membrane has a buyant density similar to pm (Stoeckenius & Kanau, 1968). The envelope of H. halobium consists of a cell wall made of two layers of protein and a cytoplasmic membrane. The cell wall shows a regular arrangement about #5 R in diameter, possibly of a glycoprotein, from electron micrographs (Blaurock et al, 1976). In the case of another halobacterium H. salinapipm, at least 50% of the cell wall protein is composed of a glchprotein. This glycoprotein is very acidic and has N-and O-linked glysidic linkages (Mescher & Strominger, 1976). These authors claim that the cell-wall glycoprotein of H. salinarium is the first one of such linkage of a glyc0protein to be found in 29 30 procaryotes. Also, the sugars of the cell wall of this halobacterium are lipid linked indicating an envolvement of a cyclic biosynthetic pathway (Mescher et al, 1976). The 2+ for normal growth, halobacterium needs about 20 mM Mg however, bacteria of aberrent shapes have been shown to grow and divide at lower Mg2+ concentrations (Henning, 1975). 2+ is needed to stabilize It seems that the divalent cation Mg the cell wall proteins. The intracellular physiology of the extreme ha10philes is dominated by the massive accumulation of K+ and Cl' ions and by the effective exclusion of Na+ (Brown, 1976). In general, enzymes associated with the cell membrane are most active at concentrations of A M NaCl or KCl, ribosomal enzymes have a specific requirement for h M KCl, and soluble enzymes have a wide range of salt optima (Brown, 1976). One interesting example of halophilic enzymes is alanine dehydrogenase from H. cutirubrum which has two specificities. As a reductive deaminase, the enzyme is fully active in the presence of high concentrations of K+, Na+ , and NH“, and also partially active with Cs+ or Li+. However, as an oxidative deaminase, it has an absolute requirement for K+. In addition, its activity increases with the temperatures upto 70 oC, although the enzyme is not thermostable (Kim & Pitt, 197?). The halobacteria are customarily grown in a complex media containing peptone and basal salts (see methods and materials) (Oesterhelt and Stoeckenius, 197A). A synthetic 31 medium containing sixteen amino acids and a variety of metal ions has been successfully used by various researchers with a few modifications (Gray & Pitts, 1976). Addition of 0.1% glycerol to the medium dramatically increases the biosynthesis of non-polar lipids e. g. carotenoids, squalenes, and retinal in the haIOphilic bacteria (Gochner et al, 1972). The colonies of H. halobium are characterized by their transluscent orange color which is due to a high content of carotenoid pigments in the cell. The harvested pellets of cells grown at a high rate of aeration are orange compared with the grayish purple cells grown at low rate of aeration. It seems that there is a branching in the carotenoid pathway leading to retinal biosynthesis (Gouchner et al, 1972). The significance of retinal biosynthesis in conjunction with purple membrane formation is discussed at the end of this chapter. The non-purple or 'red' cell membrane contains enzymes for oxidative phosphorylation and other cellular functions. By using SDS-gel electrophoresis technique, the 'red' membrane of H. cutirubrum has been shown to contain at least 25 different bands indicating as many or more membrane poly- peptides (Kushwaha et al, 1975). On the other hand, proteins of pm migrates as a single band. The molecular weight of this protein is 26,000 daltons. The lipids of the red membrane and pm are also different. In general, the lipids of extreme halophiles are very acidic(Brown, 1976). These 32 are diphytanyl ether linkages having sulfate, phosphate, or sugar polar groups. The lipids are branched and have a methyl group at every fourth carbon. However, only the pm lipids are sulfated (Kushwaha et al, 1975). A detailed discussion about red mebrane and pm lipids is presented in chapter 6. As discussed in the introduction, the dialysis of H. halobium with deionized, distilled water yields purple patches (Stoeckenius & Rowen, 1967). Until Oesterhelt and Stoeckenius (1971) and Blaurock and Stoeckenius (1971) reported the existance of the unique purple membrane protein, bR, its similarities with the visual pigment and the hexagonal lattice structure, these‘bacteria were studied only for their ability to survive in a harsh environment (Oesterhelt & Stoeckenius, 1971), (Blaurock & Stoeckenius 1971). The observation that the pm has simple chemical composition and a repeating physical structure established it as a simple experimental system amenable to several approaches in various directions. (b) Structpre of_purple membrane: The structure of pm has been studied extensively in recent years. The early freeze-fracture electron micrographs (Bluarock & Stoeckenius, 1971) show two types of structures in the cytOplasmic membrane. One shows an irregular structure and the other shows a regular structure. Comparison of these structures with isolated pm shows that the smoother, regular pattern is the pm part of the cytoplasmic membarne. X- ray diffaction 33 studies of dried and/or oriented films of pm show a regular hexagonal pattern having a P3 symmetry. An initial conclusion about the organization of the protein was that a continuous 34 A layer was formed by the protein, lipids comprising the remaining 15 R, separating pm into two domains (Bluarock & Stoeckenius, 1971). More data on the structure of pm have, however, shown that the protein-lipid distribution is different. According to newer models, which take into account the function of the bR as a light driven proton pump, spans the entire width of the lipid bilayer (Blaurock, 1975), (Henderson, 1975). These models describe pm as a symmetric unit of hexagonal P3 symmetry is composed of three asymmetric units. The electronmicrographs of tilted, unstained specimen show that there are seven 'rods' in each asymmetric unit and they are 10-12 R apart. Adjacent rods are slightly inclined to one another at various angles from 00 to 20°. These rods seem to be 30 A -#O 3 long (Henderson, 1975). The X-ray diffaction pattern leaves little doubt that these rods are «K-helices. Since bR has a molecular weight of 26,000, the -<-helices make about 70-80% of the polypeptide. Incidentally, the corrected circular dichroism spectra of pm in the peptide bond absorbance :region of 190 nm-240 nm range also show that the 1dGhelical content of bR is 75% (Long et al, 1977). The overall dimensions of bR in the model based of seven cKPhelices are 25X35X45 R. The longest of these dimension (#5 8) is perpendicular to 3b the plane of the membrane and parallel to the helices. The inner three helices from each protein almost touch each other: the outer twelve helices (four for each protein) make an outer ring which are slightly more inclined. The direction of the tilting of the outer helices is consistent with the interlocking of the amino acid side chains from adjacent °<-helices -- that is the structure is a left- handed supercoil (Fig. 3.1). It appears from these studies that the protein is globular, is almost certainly exposed on both sides of the membrane and surrounded by lipids which are arranged in separate areas with a bilayer configuration (Henderson & Unwin, 1975). The region between the inner nine helices is presumed to be a lipid bilayer as U022+ ions 2+ binds to the phosphate groups of the 2+) can bind to it (UO2 lipid region similar to Mg . This space between the three protein molecules is approximately 20 32. Also, the three protein complex of the unit cell (dimension of 62 3) is surrounded by lipids which are arranged in separate areas with a bilayer configuration (Henderson & Unwin, 1975). However, as discussed in chapter 5, the number of lipid molecules in pm seems insufficient to form a lipid domain. Packing of dihydrOphytanyl chains warrants some discussion. From the monolayer studies, it has been shown that an average fatty acid phospholipid occupoes h5-5O £2. This area causes a “.5 8 diffraction. The diphytanyl ether lipid occupies 6O 32 area, therefore, it should have a 5.1 K diffraction. However, the X-ray diffraction data of Pg g/5/Q\‘)Vi i '2: "55%;, v @AA {R n o n r 7/ ‘ , 3 ”D. 0 ‘ G (D _1 A o of __ v \ Figure 3.1 X-ray diffraction data for the structure of ' purple' membrane . Three axes of symmetry A, B, C can be seen: A seems to be the structural unit. Seven x-helices of a bacteriorhodopsin, an inner ring of nine helices and outer twelve helices can be detected (Henderson a. Unwin, 1975). 36 extracted lipids show (Blaurock, 1975) that the packing distance of diphytanyl ethers in pm is smaller, showing a h.9 R diffraction instead of 5.1 A. This indicates that the pm lipids are more tightly packed compared with their geometrically relaxed conformation lacking hindrance. (c) Comparison of bacteriorhodopsin with visual pigment (rhodopsip): The chromophore of the protein discussed above is a retinal which is covalently bound to the eyamino group of a lysine, forming a Schiff base with the protein, similar to rhodopsin (R). Hence the name bacteriorhodopsin (bR). The similarities and differences in bR and R are listed below: Similarities: (1) Membrane bound protein, undergoes reversible conforma—- tional changes upon absorption of a photon. (2) Retinal is covalently bound to an e-amino group of a lysine. (3) Schiff base is protonated in native conformation and unprotonated in bleached state. (a) Considerable red shift in the retinal-Schiff base absorption due to the interaction with protein. (5) Chromophore is all-trans in bleached conformation. (6) Acts as a photoreceptor. 37 Differences between bacteriprhodopsin(bR) and rhodopsin(R): LB 3 1) Mol. wt. 26,000, Mol. wt. “0,000, 75% protein 60% protein 2) Absorption peak around absorption peak around 565 nm. 500 nm. 3) Rigid membrane matrix Fluid membrane, protein free of two dimensional to rotate and translocate crystalline array within plane of the membrane 4) Exists in trimeric units, No such interaction exciton interaction documented between chromophores of neighbouring proteins 5) Chromophore can be 13-cis Chromophore is 11-cis in or all-trans in native bR native R 6) Chromophore is covalently Chromophore leaves the protein bound in all during bleaching intermediates 7) Proton gradient leading Overall change in membrane to ATP synthesis permeability leading to action potential In addition, the amino-acid sequence near the retinal binding site of bR is known (Bridgen & Walker, 1976). That sequence has been shown to be Val-Ser-Asp-Pro-Asp-Lys-Lys*- Phe-Tyr-Ala-Ile-Met-. The retinal binding site is at the seventh lysine which is not homologous to animal rhodopsin. 38 Also, it seems possible that the hydrophilic sequence Ser-Asp-Lys-Lys represents a link between two helices and the latter sequence is a beginning of the next helix penetrating in the membrane. However, the retinal binding site speculated by this argument does not agree with the bleaching and linear dichroism studies on investigation of the 'cavity' of retinal (see below). Amino acid analysis of bR shows that the protein is mostly in the hydrophobic region (Keefer & Bradshaw, 1977). In contrast, the rhodopsin amino acid sequence and rotational relaxation studies show that a third of the polypeptide is hydrophilic. Hydrogen excahnge studies of bR and R show that 75% of the bR peptides are hydrogen bonded (180 H-bonded, 60 free) compared with 33% of the R peptides (100 H-bonded, 200 free) (Englander &.Englander, 1977). (d) Intermediates of bacteriprhodopsip: The similarities between bR and R extend to the forma- tion of the intermediates upon absorption of a photon. The initial studies showed that a suspension of pm in basal salts saturated with diethyl ether was bleached and absorbed maximally at #12 nm (#12 nm complex or M412) in the presence of light. This photochemical reaction was shown to be accompanied by the release of a proton during bleaching due to illumination and subsequent dark uptake (Oesterhelt & Hess, 1973). Other organic solvents e. g. dimethyl sulfoxide were similarly used (Oesterhelt et al, 1973). Low temperature and flash spectroscopy have been 39 used to gain insight into the photochemical intermediates of bR. Since bR cannot be bleached using ordinary light intensities, a high intensity laser beam has been employed for such studies. Stoeckenius and Lozier (1974) have shown the presence of at least four cyclic intermediates at low temperatures. Only the first intermediate bR570 -b»bR610 requires light, the rest of them are thermal transitions. Similar work, including studies of the state of protonation of retinal-Schiff base has appeared later (Hess & Oester- helt, 1974),(0esterhelt 1974),(Chance et al, 1975), (Tokunaga et al, 1976),(Kung et al, 1975),(Lozier et al, 1975),(Lozier & Niederberger 1977). The quantum efficiency of the first intermediate is shown to be 0.4 for the forward process (Goldschmidt et al, 1976). According to these studies, at -196°C, bR570 can be converted into a roughly equal mixture of itself and a red-shifted photoproduct K610 by illumination with 500 nm light. This batho-bacteriorhodOpsin (K610) forms in less than 6 psec (Kaufman et al, 1976) and decays with a half-time of 2 psec to L550 in the dark. The next intermediate, M412 has a formation time of approximately 40’psec at room temperature. This intermediate, M412’ can be trapped in the salt-ether system by cooling to -196°C during illumina- tion.A further intermediate, 0640 is observed in experiments at 40°C but not at 0°C because of the temperature dependence of the rate constants (Dencher & Wilms, 1975),(Lozier et al, 1975). Also, an intermediate N520 is suspected because the 40 spectrum derived for M412 seems to vary with the tempera- ture of the experiment. The half-time of N520 and 0640 seems to be of the order of 5lusec each (Lozier & Nieder- berger, 1977). Figure 3.2 gives a summary of the present understanding of the photochemical bleaching. (e) Conformation of retinal in bacteriorhodopsin: (i) Extraction of the chromophore: Most of the above experiments of the photochemical bleaching were performed using a 'light adapted' bR which absorbs maximally at 570 nm (LA57O). If the pm suspension is kept in the dark, its absorption maximum occurs at 560 nm (dark adapted, DA560’ also see Chapter 7, page 115). The chromophore of LA570 has been shown to be all-trans retinal, however, there is no agreement about the chromophore of DA560. It was shown to be 1:1 mixture of all—trans and 13-cis by Pettei et al (1977) and Oesterhelt et al (1973): and excusively 13-cis by Jan (1975). Also, one of these studies have shown (Pettei et al, 1977) that the photointermediate M412 in a membrane modified by ether yields a 1:1 mixture of 13-cis and all-trans, whereas the M405 species produced by illumination of pm in 2M guanidine hydrochloride at high pH yields mainly 13-cis retinal (irreversible bleaching). It seems that the photochemical cycle of LA570 may involve an isomerization of the retinal chrom0phore from the all-trans to the 13-cis form. This is consistent with the observation that the interaction between the apOprotein and retinal seems more pronounced when the retinal is in 41 Figure 3.2a Chromophore of bacteriorhodOpsin. The chromophore all-trans retinal is covalently bound to an e-amino group of a lysine of bacteriorhodopsin via a Schiff base. PURPLE COMPLE X \ bR t‘/2=:1O psec H+ S70 . \__.// (-C ‘m Liv-'5 msec * H 0640 ('9?) K590 9/2 2’ 2 psec 1 no N [/2 — 5 msec Q 520) L550 \ (-C=N')/t‘/2°'40 psec . Mm? H /“ Figure 3.2b Model of bacteriorhodopsin intermediates. A current model showing the intermediates detected in the photochemical cycle of the purple membrane by low temperature spectrosc0py. One hydrogen ion is released and taken up again near the positions shown. 42 the all-trans form than when it is in the 13-cis form and also the apoprotein seems to impose more pronounced asymmetric constraints on the retinal in the all-trans form than in the 13-cis form ( Becher & Cassim, 1976). Also, DA bleached in the dark with dimethyl sulfoxide and 560 hydroxylamine produces only one isomer of the all-trans form, either syn or anti. In contrast, LA5 bleached with light 70 and hydroxylamine (NHZOH) produces both syn and anti isomers of all-trans retinal (Oesterhelt et al, 1973). These observations reflect changes, at the atomic level in the conformation of the protein or the retinal or both, near the Schiff base during illumination. (ii) Resonance Raman spectroscopy of retinal in bR: The extraction of retinal has a conformation close to all- trans when it is bound to the protein. The problem remains as to how to explain the difference in the spectral proper- ties of the bR intermediates upon absorption of a photon. Knowledge of the structure of retinal-protein complex and the state of isomerization of retinal induced by the conformational changes that follow the absorption of light seems to be the next step. Resonsnce Raman spectroscopy of pm and its photochemical intermediates offers an experimental approach to the analysis of the retianl conformations (Mendelsohn, 1973),(Lewis et al, 1974), (Marcus & Lewis, 1977),(Campion et al, 1977). Lewis et al have shown that in the pm complex bR57O, the Schiff base linkage is protonated and , that in the M412 complex, it is 43 unprotonated. Also, the first intermediate K590 has been shown to be protonated.(Lewis, 1976). The flash spectro- scopic results showing that the release of a proton into solution is most closely correlated in time with the forma— tion of M412 is in agreement with this. It seems from these experiments that the Schiff base may be directly involved in proton uptake and release , although the exact timing and involvement of other groups (residues) in the protein is not yet known. Kinetic resonsnce Raman data suggest (Marcus & Lewis, 1977) that there might be another inter- mediate, which contains the unprotonated Schiff base before M412 formation. Thus the release of the Schiff base proton can only be indirectly tied to the protein. The proton is eventually released forming a vectorial proton gradient across the membrane. (f) Fluorescence of the chromophore in bR: Excitation of bR in the 560-570 nm region shows very weak fluorescence around 740 nm (Alfano et al, 1976). The kinetics of fluorescence has been obtained by using picosecond laser pulses. At room temperature the life-time of fluorescence decay is estimated to be less than 3 psec (Alfano et al, 1976), however, Lewis et al (1977) estimate the life-time at room temperature to be 15 1 3 psec amd have a quantum efficiency of 2.5 X 10'“. The observation of fluorescence from the chromophore in bR is expected to provide some information about the character of an excited state that is produced before the K590 intermediate , 44 presumably that particular excited state from which K590 is formed (Lewis et al, 1977). Also, the picosecond absorption measurements suggest (Kaufmann et al, 1976) the existence of a transient intermediate between bR57o and K590. (g) Chromgphore interaction in bacteriorhodppsin: The interaction of the chromophores of neighbouring bR molecules produces an exciton coupling which can be detected by absorption and by circular dichroism of the chromophore (Heyn et al, 1975),(Becher & Ebery 1976), (Kriebel & Albrecht, 1976). Different theories have been presented by various researchers and are discussed in detail in chapter 7. The model that accounts for the most of the data is'a head to tail packing of three chromophores lying at about 200 from the plane of membrane. The results of this model are in agreement with the linear dichroism studies (King et al, 1977). The distance between two p-ionone rings, calculated using this technique, is about 18.6 R and the position of the ring is about one-third of the membrane from one surface. (h) Light induced conformational changes in bacterio- rhodppsinpinvestigaed by cross-linking technique: Cross-linking reagents like gluteraldehydes (Length= 7.5 R) or more specific reagents dimethyl adipimidate, DMA (length=8.5 X), dimethyl suberimidate, DMS (length=11.5 R) have been used to cross-link bR in the presence and in the absence of light (Konishi & Packer, 1976),(Packer et al 45 1977). This type of cross-linking can be used to determine the degree of exposure of various residues due to light induced conformational changes. These experiments indicate that the cross-linking of DMA is substantially different in the presence of light. Liposomes made from the cross-linked pm show a difference in proton transport in the dark vs light treated samples. These researchers conclude that proton may be transferred through a channel or a pore. (i) Proton gradient across the cell membrane of H. halobium: The structure of pm and the conformational changes in bR have been discussed in the preceding sections. It has been pointed out that the conformational changes in bR cause deprotonation and reprotonation of the Schiff base via a complex mechanism which is not yet understood. However, this cyclic change in the retinal Schiff base causes the vectorial translocation of a proton in the cell from the cytoplamsic side to the outside causing net acidi- fication of the medium. The net acidification of the medium can be explained as a protonmotive force (pmf) according P. Mitchell's chemiosmotic gradient hypothesis (Mitchell, 1961). Initial observations by Oesterhelt and Stoeckenius (1973) show that (i) H. halobium cells synthesize pm patches in the presence of light coupled with low oxygen supply and/or loss of other metabolites. (ii) oxygen consumption of the bacterium is reduced if it contains pm (iii) light lowers the pH of the medium in 46 which pm containing bacteria are suspended, led these and other researchers to postulate the pm as an alternate photosynthetic pigment. It was further shown (Danon & Stoeckenius, 1974) that light increases the ATP content of anaerobic, pm containing cells. Early studies using liposomes containing pm and mito- chondrial ATPase (Racker & Stoeckenius, 1974) showed that the protons can be translocated vectorially across the membrane. However, the direction of translocation of the protons observed in this study was Opposite (medium alkaline) compared with whole cells (medium acidic) due to inside out incorporation of pm. These lipsomes were shown to phosphorylate ADP in the presence of light. This study and similar recent experiments using purified ATPase from a thermophile and well defined lipids instead of original mitochondrial hydrOphobic fractions which contained electron transport chain proteins (Kagawa et al, 1977), supports the chemiosmotic theory of energy transduction in cell membranes. The liposomes generate a protonmotive force (pmf). The pmf consists of a potential gradient (AW) and a pH gradient (.A pH). The potential gradient is explained as an ionic gradient other protons and is given as follows: p: A?‘ APH .000000.0000000000000000301 A4): gT 1n [11 ooooooooooocooooooooooooBoZ ..J 1 Therefore, Ap=§T [ln[%9 + ln[%9]] ..........33 i l 47 Both the magnitude of the pH gradient ( AipH) and the membrane potential (A \P) in cells and in resealed envelopes have been estimated using lipophilic ions e. g. tetraphenyl phosphonium (TPP+), or triphenyl methyl phosphonium (TPMB+) cation (Bogomolni, 1977) or fluorescence probes e. g. cyanine dye 3,3-dipentyloxadicarbocyanine (Renthal & Lanyi, 1976) and various buffers. For H. halobium cells kept anaerobically in dark, a total pmf consisting of both Ax); and ApH components (total 130 to 150 mV, interior negative) was achieved. This pmf increased,further by 20-30 mV when the light was on. The pmf seems to be coupled to the active transport of Na+ out of the vesicles (Lanyi & MacDonald, 1976) which is used as the driving force for the light induced transport of some amino acids (see next section). The development of a Na+ gradient during illumination seems to play an important role in energy coupling. These and similar results indicate the existence of an electrogenic H7Na+ antiport mechanism (H+/Na+> 1) in H. halobium which facilitates the uphill Na+ eflux. The gradients of Na+ and H+ are thus coupled to each other such that (Na+out:;>Na+in)' These gradients seem to be capable of explaining some of the complexities of pmf (Lanyi & MacDonald, 1977). Blocking the ATP synthesing enzyme with the specific reagent dicyclohexyl- carbodiimide (DCCD) increases the pmf value to go on as high as 280 mV in the presence of light (Michel & Oesterhelt, 1976). Thus the electrochemical gradient 48 appears to be of the right size for its proposed function and can be induced by light in cell envelopes devoid of intracellular function. These measurements made by using cell-envelopes agree with those in which the pm is incorporated in black lipid membranes (Dancshazy & Karvaly, 1976),(Drachev et al, 1976), (Shieh & Packer, 1976). As discussed above, the origin of a pre-existing gradient present in the dark, in anaerobic cells may be concerned with the equilibrium of other ionic species across the cell membrane. The pmf causes phosphorylation of ADP with the ratio of 2.9 protons/ATP. Initial alkalization is explained in terms of pre-existing gredients across the membrane (Bogomolni et al, 1976). (j) Light induced active transport of amino acids: The vesicles used in the above study are also shown to accumulate amino acids in the presence of light (MacDonald & Lanyi, 1975). Subsequent studies have revealed a plausi- ble mechanism for the amino acid transport across the membrane in response to light induced electrical and chemical gredients (Hubbard et al, 1976),(MacDonald & Lanyi, 1975). Nineteen out of twenty amino acids have been shown to be actively accumulated by these vesicles in response to illumination or to an artificially created Na+ gradient. Sodium activated amino acid transport occurs in direct response to the pmf generated. Glutamate is transported only in response to a Na+ gradient. Experimen- tal evidence suggests that there are symmetrical carriers 49 that can transport amino acids equally well in both directions across the vesicle membrane depending on the direction of the driving force. (k) Bipgenesis of thegpggple membrane: The biosynthesis of pm occurs only in the presence of light and at a low rate of aeration. Pm is a simple system by which one can study membrane biogenesis. It can also be used to investigate the spatial relation between the synthesis and insertion of a protein in the membrane. In the case of bR, the synthesis of retinal and bacterio- opsin are induced separately by a low oxygen tension. Free bacterio-opsin inhibits its own synthesis. Also, bacterio- opsin inhibits retinal biosynthesis by blocking cyclization of lyc0pene, a precursor of retinal. In short, both bacterio-opsin and retinal biosynthesis is induced separately. However, they can inhibit and control each other's induction (Sumper & Hermann, 1976, a & b). Also, the bacterio-opsin formed in the absence of retinal (due to nicotine inhibition) does not form a lattice structure with addition of retinal after extraction although it can form the lattice structure if the retinal is added to metabolizing cells. This indicates that some modification of bacterio-opsin occurs in living cells after the addition of retinal. (1) Phototaxis in cell containing purple membrane: Phototactic responses in H. halobium have been studied by Hildebrand & Dencher (1975). This response can be 50 divided into two types: (1) response to a reduction in the light intensity which has an action spectrum centered at 565nm and (2) response to increased light intensity which has an action spectrum with several peaks in the ultraviolet to blue region. The function of these responses seems to be to allow the bacteria to go to high intensity of light in 565 nm region but exclude it from areas near dsmaging UV light. The first response is dependent on the presence of pm in the membrane. For more detailed information, a recent review by Henderson (1977) is recommended. 0n the other hand, for an introduction to the purple membrane without technical details, an article by Stoeckenius (1976) is sufficient. CHAPTER 4 METHODS AND MATERIALS METHODS: (a) Culture conditions for Halobacterium halobium: Halobacterium halobium strain R1 (Mutant) was used for all membrane experiments. This mutant does not form gas vacuoles (see chapter 3, page 29): therefore, contamination due to intracytoplasmic membrane can be avoided. The bacteria were stored on a 1.8% agar + growth medium (see materials) slant at 5°C and transferred every three months. Preparation of a 10 litre culture (normally a 10 litre culture was necessary for a reasonable membrane yield) was started from a single colony. A single colony either from a slant or a perti—dish was transferred to 100 ml medium in a 250 ml De Young flask. The flask was placed on a gyratory shaker by Brunswick at 150 rpm. It was illuminated by two Sylvania 10 W cool white fluorescent.lights approxi- mately 30 inches above the flask. The intensity of light at the surface of the flask was approximately 0.5 mw. The cells from the colony were grown at 37°C until they reached a stationary phase. An inoculum for a 10 litre culture was obtained by growing 500 ml culture in each of two 1 litre De Young flasks from the starter 100 ml culture. Every time, 10% inoculum was used, the initial optical density (0. D.) was between 0.05 - 0.06 at 660 nm. The 10 litre culture 51 52 ‘was grown at 37°C in a 5 gallon carboy which was placed in an incubation room. The intensity of light, provided by 12 cool white 10 W lights around the carboy, at the center of the carboy was approximately 1-3 mW/cmz. The culture was aerated by a 4 inch teflon stirrer attached to.a glass rod which was, in turn, attached to an overhead motor. The tap of the carboy was open to air and the sterility was not maintained after the 10 litre culture was inoculated. However, the absence of any contamination was ensured by the following criteria: (i) the incubation room was periodically sterilized by UV lights. (ii) a sample from the carboy at the bacterium's stationary phase was streak- ed periodically on a perti-dish: no morphological change in the colonies was observed. (iii) the growth medium was left exposed to air for a long time and streaked to deter- mine airbourne contamination and this also was negative. The initial 0. D. of the cells in the carboy was approxi- mately 0.06 and reached 0.3 at 660 nm after 150 hours. The turbidity at 660 nm was used as a measure of cell-growth, which was often verified by counting the cells in a Petrof-Hauser counter. The pm content of the cells was determined by the lysis of the cells with a 1:1 dilution using distilled water and measuring the 0. D. at 560 nm. The pm content in our preparations was parallel to the cell growth (Fig. 4.1). However, the length of the H. halobium rods decreased as the cells reached a stationary phase. 53 8 0 2.04 E .... fl __ - . - _j r 3‘ n " "-4 2 o 23 407 *E FJ-q: %' 8 2'; - v "-4 --‘ _- r-4 +3 _ I- r-4 Cl. 0) O _ r- L) p 106 I l l I 24 48 72 96 .Hours Figure 4.1 Growth curve of H. halobium Halobacterium halobium were grown in 5 gallon carboy at optimum aeration and light intensity. a. Optical density of the whole cells at 660 nm -—o—o-—o— b. Optical density of the cells ruptured with 1:1 distilled water to measure pm content at 560 m—H+ c. Cell count measured using Petrof—Hauser counter and a phase contrast microscope at 300 magnification—flfl__.__ 54 (b) Harvesting: The bacteria were harvested by centrifugation in a GS-3 rotor at 8,000 rpm for 15 minutes. The pellets from a 10 litre culture were washed twice in salts 1 (see materials) and resuSpended in 200 ml of salts 1. For the experiments where the fluidity and the phase transition temperature of cell membrane vesicles (cmv) and pm were comparedusing the spin and fluorescence probes, the harvest- ed cells were divided into three parts: one part was used for isolation of cmv, the remaining two parts were used for the isolation of the pm. (c) Isglation of H. halobium cell_membrane vesicles: The cells suspended in approximately 70 ml of salts 1 in a plastic bottle were placed in an ice bucket and liquid nitrogen was poured into the bucket such that the bacteria became completely frozen. The ice bucket containing the liquid nitrogen and the plastic bottle was stored overnight in a cold room. After thawing, the suspension became highly viscous due to the leaking of the cytOplasmic content. Two thousand kunitz units of DNAse I were added to the lysed cells and the suspension was stirred for one hour at room temperature. The cells were then dialyzed with 6 litres of salts 2 (see materials) (2 litre container with two changes every 4 hours). The dialysis tubing by Thomas Scientific Apparatus which had diameter of 5/8 inch and a molecular cutoff of 12,000 daltons was soaked in salts 2 for two hours to remove glycerol prior to use. The dialyzed cells 55 were centrifuged for 5 minutes at 6,000 rpm to remove cellular debris and then for 1 hour at 18,000 rpm in a 88-34 rotor. These two centrifugation steps were repeated until the color of the supernatant changed from deep orange to colorless or a very faint purple. A Thomas teflon tissue grinder Model C was used to resuspend the pellets. This procedure, essentially developed by Oesterhelt and Stoeckenius (1974) reportedly yields cytoplasmic membrane vesicles of H. haolpipm without the cell wall. The cmv were stable for at least 3 months according to ESR data. However, they were used within a week for the spectroscopic measurements. (d) Isolation of_purple membrane: The harvested cells were dialyzed against 15 litres of double distilled water (dd H20) in a 5 litre continuous flow dialysis apparatus. These were then treated with 4,000 kunitz units of DNAse I. The pm fraction was centrifuged for 1 hour at 18,000 rpm in a 88-34 rotor. After removal of supernatant, 5 ml of dd H20 were added to the pellets an and they wer gently shaken so that the cell-debris remained as a pellet and the pm could be decanted. The decanted pm in water was homogenized with a Thomas tissue grinder. This process was repeated until no muddy-colored cell-debris remained and the color of the supernatant changed from deep orange to faint purple. The final pellet was resus- pended in 10 ml of dd H20 for a density gradient centri- fugation. A SW Ti 41 rotor and Beckman LB ultracentrifuge 56 was used for this centrifugation. A continuous 30%-50% gradient was prepared by forming a 1 ml cusion of 60% sucrose and then adding 5 ml of each of 50% and 30% sucrose to a custom made continuous density gradient maker. Two ml of the pm sample were added to the gradient in each of six cellulose nitrate tubes. The gradients were centrifuged to equilibriate for 18 hours at 40,000 rpm yielded a dense, purple band and a diffuse, low concentration red band. The pm band in sucrose was collected with a disposable pipette after discarding the top layers. The pm in sucrose obtained from the density gradient was washed twice in dd H20 at 18,000 rpm for one hour, stored as pellets and resuspended as necessary. Similar to cmv, pm was stable for at least 3 months by the same criteria. The approximate yield of the pm was 40-60 mg/10 litre carboy. (e) Criteria for the purity of the purple membrgpp: (i) The complex medium, Bacterio-peptone from the same batch was used for all experiments to ensure that no variation in the growth medium could be experienced. (ii) The equilibrium density gradient centrifugation of continuous sucrose gradient of pm showed a single, sharp band. If the pm were contaminated by other membranes, the band would be diffuse or more than one band would have been observed. (iii) The absorption spectrum (Fig. 4.2) is identical to that reported in the literature (Oesterhelt & Stoecke- nius, 1974). As reported in the literature, the ratio of 57 anopmaonma ado CH.:30am mflaopomn one Bone meHnEme mamusm one go ezppowam soapmpomn¢ N.s mhzmam 58 ....o To.o 1|..O Tao. . 59 the 280 nm (protein) peak to the 560 nm (purple) peak is 1.9 to 1.0. (iv) ElectrOphoresis of the pm band on a SDS (sodium dodecyl sulfate) polyacrylamide gels (PAGE) was another criterion for purity. The stacking gels used were 5% acryl- amide. the separating gels were 9% acrylamide. The gels were stained with Coomassie brilliant blue (Laemmi et al, 1970). The purple membrane protein (bR) migrated as a single band. Two gels were overloaded with protein to ckeck for a small contamination: however, no minor bands could be detected which indicated no contamination. (v) To determine the protein:lipid ratio of pm in our preparation, the dry weight protein analysis of pm was performed. The vacuum drying procedure was used to determine the dry weight of pm and the concentration of protein for that sample was determined by the method of Lowry (Lowry et al, 1951). The protein:non-protein (lipid) ratio for our preparation was 78:22 (5% error) which is in close agreement with the reported value of 75:25 (Oesterhelt & Stoeckenius, 1971). (vi) The cmv were coated with 1% phosphotungstic acid (PTA), and subjected to electron microscopy which revealed the cmv to be cloed, single wall vesicles. (f) Preparation of liposgpgs: 0.1 ml of 10 mg/ml phosphotidyl choline (PC) in chloroform and 10 pl of 10 mM AS in ethanol were added to 10 m1 of liposome buffer (see materials) to make a 0.1% 60 aqueous dispersion. The molar ratio of probe:lipid was 1:50. A Bronson sonifier (Model W 185 D) was used to sonicate the dispersion at power setting 4 (meter setting approximately 60 W) in a Branson sonifier container which could be water cooled and was gas exchangeable. The temperature of circulating water was 12-1500 and the soni- cation was performed under a flow of nitrogen. (g) Purple membraneplippsomes: Closed vesicles containing fragments of pm sheets and PC liposomes were obtained by sonicating 300-500)ug of protein (pm) in pm-liposome buffer (see materials) and PC liposomes together for 10 minutes in a water jacketed sonifier container. The power setting for pm-liposomes was 2 (20 W). (h) Bleaghipg of purple membrane: The published methods for bleaching the pm vary considerably in pH and NHZOH concentration (Oesterhelt et al, 1974),(Becher & Ebery, 1976),(Bauer et al, 1976). The method described below was determined to be optimal. A fresh solution of 0.2 M NHZOH was prepared prior to bleach- ing and the pH was adjusted to 7.0 by adding NaOH pellets. A typical sample for bleaching experiments consisted of approximately 0.5 mg of bR in 15 ml of NHZOH. For illumina- tion, a three cm diameter hole was bored in a 900 W Xenon lamp housing to allow white light. The light was filtered through 8 cm optical path of 1% CuSO4 and a Corning cutoff filter 3-70 to transmit the light between 500 nm and 700 61 nm. The filtered light was collimated by a lens: the intensity of light at the sample position was 5 mW/cmz. The sample was stirred while bleaching. A fan was used to maintain the temperature of the sample at room temperature. Typically, about 30-40 hours of light exposure were needed for bleaching 1.5 mg of bR. Alternatively, 1.5 mg of bR in 5 ml of 0.2 M NHZOH, pH 7.0 was exposed to an Argon 513 nm laser for 10-15 minutes at 3 watts power. Although the temperature of the sample reached 50°C during bleaching, the degree of reconstitution of Xenon bleached and Argon bleached pm was equal. For stepwise bleaching, the absor- bance of pm at 560 nm was monitored as a function of time. One hundred percent bleaching resulted in the loss of the 560 nm peak, which could also be detected visually as the disappearance of the purple color. Bleaching of pm was also achieved by the addition of a 1:5 ratio of ether:basal salts (salts 1) added to the pm. The ether bleaching is reversible i. e. no additional retinal is needed for the reconstitution. However, the absorption maximum of the native pm is changed due to the addition of ether indicating some changes in bR and/or the membrane, therefore, ether bleaching was considered undesirable. (i) Reconstitutipn_of pppple membrane: The bleached pm was washed free of NHZOH before attempt- ing reconstitution. Typically, 10-12 pl of 1 mM retinal was sufficient for 100% reconstitution of 300 pg bR in either 62 0.01 M Tris, pH 7.0 or dd H 0. Stepwise reconstitution was 2 achieved by adding 2 ul of 1 mM 13-cis or all-trans retinal to the same concentration of bR each time (see Fig. 4.3 for bleaching and reconstitution). (j) Solubilization of purple membrane: Non-ionic detergents, e. g. Triton X-100, dodecyl trimethyl ammonium bromide (DTAB) reduced the scattering of the pm supsension significantly for the spectroscopic measurements, however, they caused a blue shift to 4-10 nm.in the 560 nm peak of the pm indicating less interaction between chromophore and protein and/or conforamtional change in the protein. Therefore, detergents were not used in this study. For reproducing some CD data (Becher & Ebery, 1976), pm was suspended in 0.3% Triton X-100 in 0.01 M Tris buffer, pH 7.0. Under these conditions, the pm became bleached overnight without exposing it to light. DTAB was used for the same purpose and was found satisfa- ctory compared to Triton X-100. (k) Gel chromoatography: This technique was used occasionally to remove free probe from the sample. Sephadex G-25 (coarse) gels were swelled in boiling water for two hours. The pm in distilled water adhered to the gel and could be removed by eluting it with 0.01 M Tris, pH 7.0. However, due to the occurance of the dilution of the sample, centrifugation or dialysis was preferred to gel chromatography. 63 II II II II I Hmsfipoh mcmuP:HHm as H mo HRCNH new: UmpSpflmeOOma ocmpnsme mamasm um:oMmHQ mo Coapmpomp¢ .o Il...ll.|.|l. Ego: on no.“ ”EMS“ manwmfi> pcm o.m mm .mommz E m.o mafims pmnommap msmanme mamhsm mo Coflpmuowp¢ .n oceansme mamasm m>flpms mo soapmaomn< .m .cOHpsvflpm:oowa use msflgomoap Mo mQOMPflpsoo msowpm> amps: :fimmoponMOMampomn ac appommm Soapmhomn< m.s magmas 64 ‘- 1.2 '00 h- 760 600 500 400 300 l '2 0 Al I‘Nlc 0-6 L- wanna 0.2 L- 65 (l) Labelling of membranes: Fatty acid fluorescence or spin probes were routinely incorporated in the membrane by mild sonication. For ESR, 8 P1.of 10mM probe in ethanol was added to 0.4 ml of 20-30 mg/ml bR sample, vortexed for 1 minute and sonicated for 10 minutes using a Cole-Parmer one quart sonicator. For fluorescence probes, 201ml of 10 mM AS in ethanol was added to 1-2 mg of bR (probe:lipid ratio of about 1:50) in 5 ml of 0.01 M Tris, pH 7.0 or dd H20, vortexed for one minute and sonicated for 10 minutes at 37°C. Different solvents e. g. hexane or chloroform were used to dissolve the probe, however, for fatty acid probes, ethanol was found to be the most desirable solvent. (m) Temperature variapipn studies: (i) A circulating water bath in conjunction with the Aminco waterjacketed cell holder was used for temperature variation between 5°C to 50°C. A mixture of ice and water in a reservoir was gradually heated by using a Bronwell Co. thermomix II, 750 W heater and pumped through the cell- holder. The temperature of the cuvette was monitored by a thermocouple thermometer with an internal reference. (ii) Bubbling of liquid nitrogen to obtain temperatures between -150°C to room temperature was used for obtaining the contraction factor and DNSA in ethanol data. A quartz dewar with a flat quartz excitation window and a one cm square Suprasil cuvette were used for these studies. The temperature of a sample was controlled by boiling liquid 66 nitrogen from a reservoir using a (5 watts, 100 ohms) power resistor and allowing the cold nitrogen gas to flow into the sample dewar at different rates. A variable temperature control accessary by Wescan Instruments, Inc. (Model 240, temperature controller) was used to lower and maintain the temperature. The probe of the temperature was inserted in the Optical dewar and the power resistor was connected or disconnected to control the boiling of liquid nitrogen according to the setting of the controller, which was previously calibrated with the temperature. A normally closed relay was used to invert the on and off cycle of the power due to the fact that an inverse relation between temperature and heater was needed (Fig. 4.4). (n) Spectral measurements: (i) Electron spin resonance (ESR): ESR spectra were recorded using Varian E 112 Century Spectrometer operated at 9.1 GHz. For fatty acid spin probes, the center of the magnetic field was set at 3270 guass. A Varian temperature control accessary was used to change the temperature of the sample, however, the actual temperature of the sample fluctuated by as much as 5°C at extreme temperatures as compared with a thermocouple. The temperature of the sample in a flat, quartz microcell was monitored by a thermocouple attached to the microcell using a Bailey thermocouple thermometer. Dry ice was stored in the reservoir and the lowest temperature of the sample (below which the aqueous membrane samples froze) was -5°C. A plot of incident 67 M .,__l. _ W A A-liquid nitrogen dewar B-power resistor C-insulated tubing for bioled nitrogen D-optical dewar E-cuvette with bulb F-thermocouple G-thermocouple thermometer H-switch for relay I-relay J-variac K-temperature controller L-probe for temperature controller Figure 4.4 Schematic of the temperature coBtrol accegsary for the temperature between 100 K and 300 K. 68 microwave intensity was set at 10 mW. Similarly, an increase in the modulation amplitude increased the gain of the signal. In this case, the fine-structure of the ESR spectrum tended to disappear as a result of the increase in gain. Therefore, the modulation amplitude was set at 3.2 G. The receiver gain was between 2 x 103 to 6 x 103. The time constant was set at 0.25 seconds and the spectra were recorded using either 4 min/scan or 8 min/scan speeds. (ii) Absorption spectra: These were recorded using a Cary 15 spectrOphotometer. (iii) Circular dichroism spectra (CD): CD spectra were recorded at the Biophysics Department, University of Michigan courtesy of Prof. Zand. A JASCO model ORD/UV-5 with a SPROUL Scientific SS-20 CD modification was used. The pm was suspended in 0.01 M Tris, pH 7.0 in a one cm rectangular cuvette. The concentration of the bR was 200 lag/ml which showed 0.D.560=1.0 on the Cary 15 and 0. D. 560:0.4 on the JASCO photomultiplier scale. The CD scale was 2 millidegrees per cm, the time constant was 4 seconds and the speed of the recorder was set at 2 which corresponded to 10 nm per minute. (iv) Emissionspectra: Routine fluorescence spectra were obtained by an Aminco-Keirs Spectrofluorimeter modi- fied with an EMI 978 IR phototube. For higher resolution, a component system was used. A 900 W Xenon lamp (powered by a Christie Silicon Rectifier (MHM-900-28) power supply), was used.as an excitation source. The wavelengths of 69 excitation were selected by a Bausch and Lomb 10 cm grating blazed at 3000 R in a B & L 500 mm monochromator. A TRW sample holder with ccllimating lenses were used for aligning the excitation and emission components. The emission monochromator was a 750 mm Czerny-Turner spectro- meter (Spex 1711-11) in conjunction with a Princeston Applied Research PAR-lock-in amplifier (model HR-8) and an EMI 9558 QA phototube. The power supply for the phototube was a Fluke 412 B which normally operated as 1100 V. The signal, with a chopper as a reference was connected to a Bristol Dynamaster recorder. The resolution of this system was 2 nm. (v) Quantum yields: Quantum yields for DNSA in various solvents at room temperature were obtained by utilizing a double beem quantum yield instrument interfaced with a PDP 11 computer (Holland et al, 1973). Quantum yields for DNSA in ethanol at different temperatures and for AS in various solvents were obtained by using an Aminco-Keirs instrument and comparing the measured areas of the sample and the reference. The O. D. of the sample and the reference was kept less than 0.2 for all wavelengths and adjusted to be equal at the excitation wavelength. The method of comparing the areas of the spectra was similarly used to measure the changes in the abosrbance as a function of temperature to obtain the contraction factor. Scattering due to the optical dewar was compensated fro in both cases 0 70 (vi) Nanosecond time resplved fluorescence decay measurements: The nanosecond fluorescence decay curves were obtained with a single photon counting, time resolved spectrophotometer built in our laboratory and described elsewhere (Avouris et al, 1974). This instrument consists of a deuterium nanosecond flash lamp, a 1P28 phototube to start the signal. The sample holder, an emission monochro- mator (B:& L 500 mm, similar to the one described in the emission spectra), a high gain (108) and low noise photo- tube Amperex 56 DUVP which is capable of amplifying a single photon into an electric pulse to stop the signal, a time to amplitude converter Ortec 475, a multichannel analyzer Nuclear data 1100 to collect and store the data: also to transmit the data to the computer, HP 7200 A plotter etc. (Fig. 4.5). Interference or band-pass filters were used for excitation. Various computer programs are used to calculate the life-time of the fluorescence decay. A deconvolution analysis developed by Ware (ware et al, 1973) and tripex analysis (Ware W. R., personal communica- tion) are found satisfactory for multicomponent decay. Details of various computer programs to calculate the life-times corrected for lamp shifts and for photmultiplier response are in preparation (Kennedy & El-Bayoumi, in preparation). 71 "‘28 LAMP SAMPLE 56 DUVP BASE 21.7 llFETIME :7") APPARATUS "M ‘ o: LAY c A L I 3. ——dl I 0 use use .flv sun SIOP T A c L___, COUPLER H PLOT‘IER Figure 4.5 Nanosecond Time-resolved Spectrofluorimeter. 72 MATERIALS: Chemicals: N-oxyl-4: 4'-dimethyl oxazolidin 5-keto stearic acid (5-NS) and n-oxyl-4z4'-dimethyl oxazolidine 12-keto stearic acid (12—NS) were purchased from Synvar Corp., Palo Alto, California. 12-(9-anthroyl) stearic acid (AS), dansyl sulfanoamide (DNSA), 13-cis retinal, all-trans retinal, DNAase I from bovine pancreas were purchased from Sigma. 1,6 diphenyl 2,4,6-hexatriene (DPH) was purchased from Eastman. Diphenyl anthracene (DPA), anthracene- 9-carboxylic acid (ACA), and naphthalene were purchased from Aldrich. Solvents: Ethanol, n-propanol and diethyl ether were redistilled in our lab. 3-methyl pentane (3-MP) was purified by the method of Potts (1952) modified in our laboratory and reported by Avouris (Avouris, 1974). Spectrograde glycerine was purchased from Aldrich. Spectro- grade EPA was purchased from the American Instrument Company. Spectrograde methanol was purchased from Mallin- crodt. Spectrograde benzene, cyclohexane, hexane, and 2-methyl butane (isopentane) were purchased from Matheson, Coleman and Bell (MC/B). Growth medium: For 1000 ml of medium: NaCl....................... 250.0 gm MgSO#.7H20................. 20.0 gm trisodium citrate.2H20 .... 3.0 gm KCl ....................... 2.0 gm ca01202H20 ......OOOOOOOIOO OOng 73 Ten gm Bacto-peptone (Bacteriological-technical by Difco) were mixed separately with 100 ml of dd H20 and was auto- claved. The salts were dissolved in dd H20 to make an 850. ml volume and were autoclaved separately. After autoclaving, the peptone was mixed with the salts: the pH was adjusted by adding 5 ml of 0.5 N NaOH to 7.4: then the total volume of the medium was made to 1000 ml with sterile, dd H20. Buffers: Salts 1: These are 'basal salts': salts that are listed in the medium without peptone. .s_a_L_1_t_s_2.= 1.0 M NaCl 20 mM MgSOu.7H20 10 mM Tris-Cl pH 7.4 Lipcsome buffer (P01): 0.1 M NaCl 0.05 M Tris-Cl pH 8.0 Purple membrane liposome buffer: 0.15 M KCl 0.01 M Tris-Cl pH 7.0 CHAPTER 5 PHOTOPHYSICS OF A POLARITY PROBE (DANSYL SULFANOAMIDE) AND A PACKING PROBE (ANTHROYL STEARATE) ' (a) Introduction: As discussed in chapter 2, the polarity probes e. g. DNSA, ANS, AS are weakly fluorescent in solvents of high polarity. As the polarity of the medium decreases, the quantum yield of fluorescence of the probe increases and the “max shifts towards higher energies. These effects are interpreted in terms of the solvent relaxation of the polar solvent around the probe molecule. The degree of the solvent relaxation depends upon the ability of the solvent molecules to reorient themselves around the probe molecule during the life-time of the excited state. TherefOre, in a highly viscous, polar medium, even though the dipole moment of the solvent molecules is high, the solvent relaxation does not occur during the life-time of the excited state giving rise to an emission similar to that in a non-polar solvent (solvent which has a lower dipole moment). In this chapter, the medium effects (polarity and rigidity) on the lumini- scence properties of two probe molecules, dansyl sulfano- amide (DNSA) and anthroyl stearate (AS) are studied in detail. In order to differentiate between the properties of DNSA in a polar and viscous versus in a non-polar and fluid medium, the effect of temperature on the emission 74 75 maximum, fluorescence quantum yield and life-time of DNSA in a polar solvent is studied. This study required the measurement of solvent densities as a function of tempera+ ture in order to compensate for absorption changes due to the solution contraction. These contraction factors were also obtained for three other widely used organic solvents. Photophysical studies of probes is divided intc.three parts: (b) Contraction factor: Temperature dependence of the fluorescence quantum yield can be utilized to gain significant information about the electronic transitions as well as the intramolecular interactions in the excited states. However, caution must be exercised while interpreting the change in the quantum yield of a fluorescent molecule upon changes in the tempe- rature. The fluorescence quantum yield (Pf(T) as a function of temperature is determined relative to its value of ¢°f(T) at room temperature by the equation (Mantulin & Hubert, 1973) is given by: jF(D,T)d9 D n2(T) 01m) =¢°f(T) ........ _- -——-_-_ ......... 2-- .....5.1 jFo( V,T)d>' D(T) no where F is the corrected fluorescence spectrum D is the cptical density of the solvent n is the refractive index of the solvent Therefore, to obtain a true increase in the quantum yield of fluorescence at a lower temperature, the change in the density of a solvent must be compensated for. The ratio of Do/D(T) in the above equation compensates for the 76 temperature dependence of (i) the density of the solvent and (ii) the bandwidths, spectral shifts and transition probabilities fo the absorption peaks. The ratio of Do/D(T), termed as the 'contraction factor' can be obtained by the measurement of the absorption spectrum of a molecule in the same spectral range as the excitation of most aromatic or conjugated linear probe molecules. We studied this contraction factor quantitatively by determining the change in the absorption spectrum of 9-10 diphenyl anthra- cene (DPA) in different solvents as a function of temperae. ture. The contraction factor was experimentally determined as follows: Absorption spectrum at room temp. Contraction factor = ---------------------------------- Absorption spectrum at lower temp. Area of absorption curve at room temperature Area of absorption curve at a lower temperature Assuming that the extinction coefficient of DPA is independent of temperature, the increase in the intensity of absorption by lowering the temperature was solely attributed to the ratio Do/D(T) for that particular solvent. As stated earlier, the contraction factor'was obtained for the following solvents: (i) ethanol (ii) n-propanol (iii) 3-methyl pentane (3-MP) and (iv) methyl cyclohexane (MCH): is0pentane (IP) in the ratio 3:1. Also, two probe molecules, DPA and naphthalene were used to obtain the contraction factor for ethanol to show that the 77 contraction factor is independent of the properties of a solute (Fig. 5.1). It can be seen from the equation 5.1 that the correca tion in the refractive index is also necessary to obtain the total correction factor Efflr). The 3RT) is given by }(T)=--]-39 ----- . ------ 2--- 52 We have obtained Do/D(T) (contraction factor) by the experimental procedure, that factor can be used ot determine the refractive index at various temperatures by using one of the two equations given below: (i) --------- = constant ....................5.3 (ii) ----------------- = (T) ................... 5.4 b - n(T) + 0.4 The plots from these two equations showed that the equation 5.4 was more satisfactory as the deviation in the linearity of the experimental points was lower. The constant in the equation 5.4 was obtained by using known values of the refractive index obtained by the above equation for ethanol, n-propanol and 3-MP at various temperatures is shown in Fig. 5.2. As mentioned earlier, by combining the contraction factor obtained from the absorption spectra and the refrace tive index obtained by the equation 5.4, the total 78 0.95 0.90 EACYOI CONVIACI’ION 0.8 1‘ " l ‘ 1'30:: ,3 :50 125 nun-tum“ "K Figure 5.1 Contraction factor Contraction factor for different solvents as a function of temperature. Solute used for all solvents, diphenyl anthracene (DPA). for ethanol, another solute, naphthalene was used. a. propanol—o—o— b. 3-methyl pentane—W— c. methyl cyclohexane:isopentane (3:1) —4L—qc—— d. ethanol-DPA—o——o—, ethanol-naphthalene—I—I— I-CI' I .45 X H D 51.40 U 2 5 ‘ E U ‘10:, - 79 J l I l 100 150 200 230 "annual: “I: Figure 5.2 Refractive index of various solvents Refractive index of the solvents a. ethanol —a-—a— b. propanol—o—a— c. 3-methyl pentane._45_45__ Equation used: (n-1)/d = constant 8O correction factor 3*(T), to be applied to tbe corrected fluorescence is given by the known values of the equation 5.2. The Ef(T) is made available for the following solvent solvents: (i) ethanol (ii) n-propanol and (iii) 3-MP (Fig. 5.3). The results of this study are applied to correct the quantum yield of DNSA in ethanol at different temperatures. The contraction factor should prove valuable in studying emission and absorption properties of molecules as a function of temperature in these solvents. (c) Dansyl sulfanoamide: Dansyl (2—2 dimethylamino—naphthalene—6-)sulfanoamide (DNSA) is a "polarity probe" which is widely used to deter- mine the nature of its binding site and changes in its fluorescence parameters due to conformational changes in biological macromolecules and membranes. It has been shown that DNSA forms a highly fluorescent complex with bovine carbonic anhydrase at its active site and upon binding, its fluorescence properties enhance dramatically (Chen & Kearnhan, 1967). DNSA and other dansyl probes which have a fluorescent 2-amino-naphthalene-6—sulfonate group attach- ed to a functional group such as an amide, chloride, amino acid, cadaverine etc are used in the biological systems according to the specificity of their functional groups. However, the fluorescence properties of these molecules should be studied in the simple solvents in greater detail to derive more information about biological systems. The fluorescence properties of some dansyl derivatives have 81 OI ..u o:— I VAC 0.85 '- l 1 l I so :00 m. zoo 25 no tmnnwu 'x Figure 5.3 jfactor for various solvents 3- factor for various solvents were determined by combining the contraction factor and the refractive index at individual temperatures a. ethanol—u—g— b. propanol__o_o_ c. 3-methyl pentane—....— 82 been measured in mixed solvents, one polar and other non- polar (Li et al, 1975),(Seliskar & Brand, 1971). The dipole moment of this molecule increases upon excitation, thus the relaxation of the solvent cage leads to different emission maxima depending on the polarity of the solvent as shown in Fig. 5.4. We have studied the fluorescence properties, e. g. an emission maximum, fluorescence quantum yield and life-time of fluorescence decay of DNSA in solvents of varying polarity. The results are shown in Table 5.1. One may observe that as the 'polarity' of the medium increases, the quantum yield (¢f) and life-time ('z'f) decreases and the wave— length maximum (Amax) increases. Degassing by bubbling dry nitrogen leads to an increase in the value of (bf, and ’c’f. without a change in “max' This is interpreted in terms of oxygen quenching of fluorescence via intersystem crossing enhancement mechanism. That is, oxygen enhances the spin-orbit coupling hence increasing the rate constant of intersystem crossing to triplet state leading to an observed fluorescence quenching. The effect of oxygen quenching is particularly significant in benzene medium. It can be seen that the 'rf of DNSA was 19.1 nsec in n-propanol but decreased gradually to 16.0 nsec in methanol and then dramatically to 2.2 nsec in water. Similarly, the emission wavelength maximum of DNSA changed from 495 nm in propanol to 505 nm in methanol and to 550 nm in water. Interaction of water with DNSA leads to the large observed 83 chapmthEop Econ pm cocaocoh ohm mupoomm .Aonma .wsaczmmnam Scam cousccpmomv mesmPHom msowmm> Ca Hcm wacwnm> CH «mzm mo NEHNME Cowmmfiem d.m oh:Mwm nan: x aboaiaci>ls> on as on an O < O . if M ... . mom « ... U n . m ’ a w r , loo. tn wwym {Zach w t O m M O m ©© 2 n5\ Into :,0: - 3 .8: 0mm H No ..wswmmmmgc Hopmm cam 98ch cousmMoE onsvmhmmEov Soon we mpCoPHcm msofiumdr a...” .«mzm S muoH .8 A 5:76 ..wnv LN; mowvnomoun Cowmmwem mpCoPHcmnmscwnm> Ca aom omm omc Hocmnpmz Hocmnpm Hocmmohm. mcmucmm Rheum pcm>aom 85 shift of the emission maximum and life-time is due to an expected enhancement of radiationless transition via the triplet state (SlanmPT) and/or directly to the ground state (SIMSo)' The fluorescence life-time ( ff: 6.2 nsec) and the quantum yield ( Q&F=0.11) of DNSA in D20 is more than double that in H20. This is interpreted in terms of a deuterium isotOpe solvent effect which limits radiationless transitions. It is interesting to note that in benzene, the emission maximum occurs at 475 nm which is at a longer wave wavelength than ether (470)nm . Benzene has a lower dielectric constant than ether and is non-polar. This may imply specific interactions of DNSA with benzene. Natural fluorescence life-time of DNSA for all solvents is calcu- lated from the experimental (bf and 'rf such that To = ---------- 00000.0...0.000.000.000000505 The ‘Z: is given by Strickler-Berg equation as follows: 1 811:: 2303 nf3 < ’3>-1 e dy 6 ----- = --------------- 9f J---g-'O O O O O O .5. To C2 N na Av which is modified to 1 -1 e du --.E-- = 2.88 x 10"9 n2 VEB> Av f"';,"' 5'7 o . 86 -3 where n is the refractive index of the solvent andAv is the reciprocal of the mean value of the frequency over the fluorescence spectrum. The quantity -15 ’E c is calcualted as the radiative transition probability from the lowest excited electronic state to the ground state. The trend of 7%8 is similar to -;6 which means that the m 0 changes in the fluorescence energies and refractive indices of the solvents accounts for the observed radiations. The emission parameters of DNSA obtained in ethanol at different temperatures (3000K to 770K) are shown in Fig. 5.5. As the viscosity of ethanol increased upon lowering the temperature, 'rf decreased and).max also decreased in a more or less parallel fashion. Thus, in ethanol at room temperature (fluid medium), “if = 18.6 nsec, 7\ = 500 nm, max while at 770K, (rigid medium), if = 13.5 nsec and xmax = 450 nm. It is difficult to distinguish between a polar-rigid medium and a non-polar fluid medium on the basis of 'z’f and “max measurements alone. To distinguish between these two situations, one requires the measurement of nanosecond depolarization decay of the probe molecule. Longer relaxation times are expected in the rigid media. (d) Anthroyl stearate: Anthroyl stearate (AS) is a fluorescence probe designed to investigate the hydrocarbon lipid region of an artificial or a biological membrane. The fluorescence moiety 9-anthroic carboxylic acid is covalently linked to 87 ‘ Q l mocondl =1 l T E 3 I P a O 1 36'- 26- , 7;... nsocomlo 23- s 2:- ? .1 g zo- fl 3" ‘9.- J l 1 fi :4 1 ‘0 100 ‘50 200 250 300 350 nnnutuu 'K Figure 5.5 Emission properties (‘8 f, ¢f, 't°m and 1‘1?) of DNSA in ethanol at various temperatures. 88 stearic acid at the twelfth carbon position to anchor the probe in the membrane. Since the report of its synthesis in 1970, (Waggoner & Stryer, 1970), AS has been widely used for the determination of (i) the orientation in the model membrane (Yguerbide & Stryer, 1971), (Badley, 1973) (ii) the effect of cholesterol on erythrocytes (Vanderkooi et al, 1974) (iii) the state of energization of the chloro- plast membranes (Vandermeulen & Govindjee, 1974) and many other biological membranes. In our study, we are focusing our attention on an excited state intramolecular processes of AS that explain its emission properties. This probe undergoes a geometric relaxation in the excited state such that the plane of the carboxyl group becomes planar with respect to the anthracene moiety: the extent of such relaxation depends on the viscosity of the medium. In addition, the dipole moment of AS molecules increases upon excitation leading to solvent cage relaxation depending upon the solvent polarity. The emission spectra of AS in different solvents were obtained (Table 5.2). Our results showed slightly larger shifts in the emission maximum than previously reported (Waggoner & Stryer, 1970). The shift of the fluorescence maximum from 450 nm in 3-MP to 475 nm in methanol, both at room temperature, is due to the relaxation of polar metha- nol molecules around excited state AS molecules. The emission maximum of AS in ethanol at room temperature is 468 nm. 89 Solvent 3-MP Methanol Ethanol Hexane Glycerine 'Z2' 6.8 1.8 3.1 6.5 8.5 nsec Q. 0031 00061 Oo18 0037 ‘ 9kmax 450 475 468 450 453 nm Table 5.2 Emission properties of AS in various solvents at room temperaaure. ex: 365 nm. 90 In order to understand the effect of the increase in the viscosity of a polar medium, AS was dissolved in glycerine. Although glycerine is more polar than ethanol (dielectric constant of glycerine is 42.5 compared with that of ethanol is 24.55), the emission maximum of AS in glycerine was at higher energies (453 nm) than in ethanol (468 nm). Also, the life-time of AS was higher in glycerine ( Q& = 8.5 nsec) compared with that in ethanol ('rf = 3.1 nsec). These results are interpreted in terms of the higher viscosity of glycerine (1.490 cp) compared with ethanol (1.2 cp). In order to further investigate the effect of the viscosity of the solvent on the AS emission, the fluores- cence spectra of AS in 3-MP was obtained at room tempera- ture and at 770K. The emission maximum of AS in 3-MP at room temperature was at 450 nm, that at 770K was at 427 nm (Fig. 5.6). This shift to higher energy is due to the increase in the viscosity of the solvent indicated an excited state intra-molecular geometric relaxation which occurs in the fluid medium. Experiments were performed to investigate the emission properties of AS in a polar medium (ethanol) both at room temperature and at 770K. The emission maximum of AS in ethanol at room temperature, as stated earlier, was at 468 nm, that at 770K was at 407 nm (Fig. 5.7). Emission at such a short wavelength (407 nm) in rigid ethanol cannot be solely due to the prevention of the solvent molecule relaxation in the excited state. Our 91 ‘OG 421'an 4450 nm 80-- y .— 3 2 II p- E 50.. U U 2 III U 0: III fl ‘0 3 . 40- a II E. cl < t, I O z 20- o 1 1 1 1 1 4 1 l 1 1 1 1 I 1 1 1 l 400 450 500 350 Figure 5.6 Emission spectra of 10'5 M AS in 3-methyl pentane (1) room temperature and (ii) 77'K. “ex: 365 nm. 92 ‘0’ M“ 460 Bill INVENSIYY 60- ‘OP uonquuzzo nuouscmc: 400 450 500 550 Figure 5.7 Emission spectra of AS in ethanol at (1) room temperature and (ii) 77'K. fiex=365 nm. 93 interpretation of these results is in terms of a ground state configuration of AS in which the anthracene ring is more out of plane with the carboxylic group in ethanol compared with 3-MP. Hydrogen bonding of ethanol with the carboxylic group constrained by the fatty acid chain faces such configuration. Thus, in rigid ethanol, the emitting species is essentially anthracene, hence more structured blue fluorescence (407 nm) is observed. In order to investigate the effect of fatty acid chain on the emission properties of anthracene carboxylic acid, ACA (without fatty acid chain), were measured. Fluorescence life-time of AS and ACA was equal (3.2 nsec) at room temperature, at 77°K, life-time of AS was 11.2 nsec, that of ACA was 12.1 nsec. The emission maximum of ACA was at 465 nm at room temperature, that in rigid ethanol occured at 422 nm (Fig. 5.8). In this case, the constraining effect of the fatty acid chain is absent and a relatively more planar configuration is possible. 94 ‘00 "20 m V r— ‘60 MI! - > .. A 2. a: 2: U k F- —n E I U ‘21- .— U U a II I OH‘ .— 3 .1 S D ,_ _ II E 2‘ 1I'- ._ I *0 I: 7 _- o A l :4 J l 400 450 500 550 . ‘5 0 t F' re .8 Emiss10n spectra of 10 M.@CA in ethanol a lgu 5 (1) room temperature and (ll) 77’K 9ex=365 nm CHAPTER 6 FLUIDITY STUDIES IN PURPLE MEMBRANE AND CELL MEMBRANE VESICLES USING ELECTRON SPIN RESONANCE (ESR) AND FLUORESCENCE PROBES (a) Introductipp: In this chapter we have combined spin and fluorescence probe techniques to make a comparative study of the flui- dity of pm and cell membrane vesicles (cmv). Fatty acids labelled either with a spin probe or a fluorescence probe are incorporated in pm and cmv to investigate the micro— viscosity, phase transition temperature and lipid-protein interaction in these membranes. Our studies show that the phase transition temperature of pm is higher than that of cmv and does not depend upon the presence of Na+. The change in the fluidity of pm above the phase transition is small compared with the case of cmv. Energy transfer experiments suggest that the rigidity of pm is due to the fact that most of the lipid molecules interacting with bacteriorhodopsin (bR). The probes used are shown in Fig. 6.1 (b) Fluidity and phase transition temperatures of purple membrane and cell membrane vesicles using spin probes: A typical ESR first derivative spectrum of pm at room temperature labelled with 5-NS is shown in Fig. 6.2. The hyperfine splitting (2T/7) of 5-NS in pm is 62.6 G at room 95 96 O N-PO 5-nitroXyl stearate (5-NS) CH3-(CH2)5 ;C< (CH2)|0— COOH O N '90 12-nitroxyl S ( $3515? . 1249-2222521? (AS) Figure 6.1 Structure of electron spin resonance and fluorescence probes for the fluidity studies 97 .w ommm pm wow caoflm .Nmo H.m pm pom was mesmsvopm o>mchcflz . a mm 2a sampoca sa as s om as as s.o op cases some mz-n as as so as ea .m::m cums wwaaopma ocmanEoE oH use on» Me sapwoomm o>wpm>fiaoc pmnwm mmm awoanhp < Em Ca mz:m mo Esupoomm mmm Hmcwmhv ¢ N.m oHSWMm $200 on Co. on ON 0. o 0.: ON. on: ow- on... _ _ - q q _ . _ - :OO: Kusuawl Iona 98 temperature. This value of 2T// is one of the highest for biological membranes (Chignell & Chignell, 1975).which shows extreme immobilization of the spin probe. The 2T// of the cmv, although smaller than that of pm, is also large, 61.2 G, reflecting a high viscosity of cmv. The plots of 2T// of 5-NS in pm and cmv against temperature are shown in Fig. 6.3. The thermotrOpic phase transition of 5—NS occurs at about 31°C for pm and at about 22°C for cmv. The phase transition temperature of pm, 31°C, is slightly higher than the published value (Chignell & Chignell, 1975). Although the 2T// of 5-NS is higher than that of cmv at all tempera- tures, the change in the slope of the 2T/7 with the tempe- rature (above the phase transition temperature of pm) in cmv is much larger than in the case of pm. The phase transition temperatures reported here probably represent a transition from gel + liquid crystalline to liquid crystal- line lipid regions (Lee, 1975),(Oldfield et al, 1972),( (Shimshick & McConnell, 1973). The small change in the slope of the 2T// of the pm below and above the phase transition reflects the restricted mobility of the pm lipids in the liquid crystalline phase. The pm suspended in deionized distilled water gives the same phase transi- tion temperature as that suspended in salts. A sample of pm was dialyzed with 10 mM EDTA, redialyzed with distilled water, then it was washed and resuspended in either 1 mM or 100 mM Mg2+. The 217/ of 5—NS in the pm at 37°C remained unchanged for all three (0, 1 mM, 100 mM) concentrations 99 chapmpomsov mo chvoca m mm .26 33 EB Em AC CH mZIm mo\\em mo 903 m6 mpsmfim 0.. 828369.. on O? On ON 0. O — — q q A 4 o 1. o 100 C O a a o "I. o O I D o o D . . Loo m .. . . UO-n . C C o o o o I . re. Inc /o o - P p — b _ 100 2 of Mg + for one set of experiments. Our data indicate that the monovalent cation Na+ and possibly divelent cation Mg2 + do not change the value of 2T// of 5-NS in the pm at 37°C. It appears that these cations do not influence the organization of the lipids in the pm at growth temperature. It has been shown that in case of T. acidophiig, the addi- 2+ raises the temperature of the tion of divalent cation Ca gel to gel + liquid crystalline transition (Weller & Haug, 1977). For pm, that transition may occur at a low tempera- ture and thus could not be detected. The comparatively low value (22°C) of the phase transition temperature in cmv indicates that the spin probe is incorporated predominantly in regions other than pm patches which are more fluid than pm, particularly above the phase transition (Oldfield et al, 1972). The cmv for the above experiments were isolated from cells grown at the optimum conditions for the biosynthesis of pm. The extent of pm was approximately 40% w/w with slight variation depending on the preparation. To deter- mine whether the presence of pm changes the phase transi- tion temperature or not, the cells were grown at a low light intensity and a high rate of aeration to inhibit pm biosynthesis. These cells were lysed and assayed spectro- photometrically. They did not contain the characteristic pm band at 560-570 nm. The thermotropic phase transition of the vesicles isolated from these cells occured at about 19°C compared to 22°C for the cmv containing the pm. 101 In order to investigate the rigidity of the hydrocarbon region of the cmv towards the center of the bilayer, they were labelled using 12-NS. The thermotropic phase transi- tion of 12-NS incorporated in the cmv occurs at about 18°C compared with 22°C for 5-NS as shown in Fig. 6.4. The relatively small change in the phase transition temperature using 5-NS and 12—NS indicates that the rigidity of the hydrocarbon region of the cmv extends at least upto 12 th carbon position. It is well established that H. haipbium polar lipids are mainly diphytanyl ether analogs of phospholipids and glycolipids (Kates, 1972),(Marshall & Brown, 1968),(Plachy et al, 1974). These phytanyl ethers contain a methyl group at carbons 15, 11, 7 and 3 from the ether linkage. Model studies indicate that the branching methyl can pack rather well in all-trans configuration if the lipid chains are tilted to about 20° from the normal or if the lipid mole- cules are staggered (Plachy et al, 1974). The resulting closer packing extends the rigidity further down the hydro- carbon chain compared with the saturated, unbranched situa- tion. The closeness of the phase transition temperature of 12-NS (18°C) and 5-NS (22°C) in cmv may indicate regions of similar rigidity between the 5 th and 12 th carbon due to closer packing. On the other hand, as proposed by Plachy et al (1974), the nitroxyl radical of the 12-NS molecule may be in the glycerol region at low temperature and in hydrocarbon region at high temperatures. If this is the 102 62... I I l l I I I l ‘- f5l- _ 6C)“ - 8 '3 C! O .. '8 °C Prtfia- l N H 5E3‘ a 57’ l ‘ I l J 1 1 '5 0 5 l0 IS 20 25 30 3‘ 40 Temperature ’C Figure 6.4 Plot of 2T// of 12-NS in cmv as a function of temperature. 103 case, the 12-NS results may not mean extended rigidity upto the twelfth carbon. Although this point cannot be resolved conclusively, our fluorescence studies make us feel that the rigidity extends upto the twelfth carbon. (c) Fluidity and_packing,of_purpip_pembrane and cell membrane vesicles using a fluorescence probe, anthrpyl stearate: The pm, cmv and, as a control, phosphatidyl choline vesicles (pcv) were nibelled with AS in pcv suspended in liposome buffer. The emission maximum of AS in pcv occurs at 446 nm. The emission maximum of AS in cmv occurs at 437 nm and has a slight shoulder at 417 nm. Finally; the emission maximum of AS in pm occurs at 435 nm and has another peak at 417 nm (Fig. 6.5). The life-time of decay of fluorescence of AS in pcv is 12.6 nsec (Vander koci et al, 1974),(Waggoner & Stryer, 1970). The decay of fluorescence of AS in pcv can be fitted by a single exponential. On the other hand, there are ' three components of the fluorescence decay of AS in the cmv and in pm. The shortest one is attributed to the scattering due to the size of the membranes. In order to keep the probe:lipid ratio below 1:100, the concentration of the membrane is usually 20-30‘pg/ml, which causes some scatter- ing. The other two components are '21 = 3.6 nsec and '82 = 10.0 nsec for the cmv. They are '21 = 3.2 nsce and :22 9.1 nsec for the pm. The (t1 component of the decay is attributed to the AS probe molecules that undergo 104 VT :oo , E9 Normalized Intensity g; L—A—J—i J 11 _ 380 400 450 500 Ann: Figure 6.5 Emission spectra of 10'5 M AS in (a) pcv in 11 osame buffer, (b) cmv in salts 2, and (a pm in distilled water. ‘ 105 energy transfer by Ffirster mechanism (Farster, 1951) to the chromophore of the bR or to the carotenoid pigments in the cmv. Such energy transfer is possible due to the spectral overlap between AS emission and the absorption of bR or carotenoid pigments. To verify that the origin of the '81 component in the pm as being due to fluorescence decay of AS involved in the energy transfer, the pm labelled with AS was bleached. The Spectral overlap between the emission of AS and the absorption of the bleached pm (oxime absorbs at 355 nm) is much less than with the absorption of bR in the native pm and hence the energy transfer is expected to be less efficient in the bleached pm. Upon bleaching AS labelled pm, the emission maximum of AS remained at 435 nm, however, the fluorescence intensity increased almost 100% and the long component '32 of the fluorescence decay became more dominent compared with the case of native pm. These results are interpretted in terms of an efficient energy transfer from AS to bR. The decay of AS fluorescence in E. coli outer membrane vesicles (ecomv), which is another rigid membrane (the rigidity of ecomv is similar to that of cmv as measured by 2T/y'of 5-NS, Gupte & McGroarty, unpublished results), but where no energy transfer is expected, has been measured. As expected, the contribution of the long component '12 to the fluorescence decay of ecomv is much larger than in the native or bleached pm (Fig. 6.6). The study of the fluorescence properties of AS in the 106 3 , 1 If», T l 1 ~ 1 o’. b .0 o C *. L \ .:0 .o.“o » .070 a... o . m. . g 'o ‘0. P .o .0 \‘oo. .0. . ..o "‘ p Q .. .... ...“. . ‘ ... ....s P 9. 0. “at” . ‘ '. 5. ...9 . N .0. .v . N ’ ... ... s“. .0 t 0 ~ ‘ . . o I. '0. '0. 3 .. u: “a. v- ' ' ': .... . "i 2 P ‘ N ‘0 ~ on ' ‘ ..- P o P .0 . o “.0? P 0 .05 o ‘0 ...r.. P D " 'o.‘ 0‘: - L 0'... 50'. "' ~ :' C .0 8 ” ~’ “N- P . I . - u. O... f. .. . . .0 .... P . E. o ..v. 3 - .-.. o '0' .... P- ... . d ' .. . fl-o. .... P o . c 00 ‘ l" o P .0 ' ’ o. .... P i. o. ‘ P '. 00'. > I o :0 c 1. l J 11 1 IO 20 30 40 50 lkmaumumh Figure 6.6 Decay of AS fluorescence in various membranes Decay of AS fluorescence in various membranes was obtained by nanosecond time-resolved spectro- scopy. A 365 nm interference filter was used for excitation and decay was monitored at the indivi- dual emissian maxima. The decay was observed in the following membranes: (a) pcv, (b) ecomv, (c) bleached pm, (d) cmv, and (e) pm. 107 pm and cmv and coupling these results with those obtained using spin probes 5-NS and 12-NS strengthens our previous arguments and adds to our understanding regarding the structure of these membranes. As discussed in Chpater 5, the AS probe undergoes a geometrical relaxation in the excited state such that the plane of the carboxyl group becomes more planar with respect to the anthracene moiety: the extent of such relaxation depends on the viscosity of the medium. The emission maximum of AS in fluid hydrocarbon medium occurs at 450 nm, whereas in rigid hydrocarbon medium (770K) the maximum shifts dramatically to 427 nm. In addition, the dipole moment of AS molecules increases upon excitation leading to solvent cage relaxation dependent on solvent polarity. The fluorescence wavelength maximum of AS shifts to lower energies and the fluorescence life-time decreases as the polarityof the medium is increased. The fluorescence maximum of AS in pcv shows that the probe is in hydrocarbon environment (Waggoner & Stryer, 1970). It has been shown that the emission maximum shifts to higher energies as the temperature is lowered (Vanderkooi, 1972). This indicates that the increased rigidity of lipid environment at lower tempera- ture shifts the emission maximum of AS to higher energies. In our experiments, the AS emission shifted from 446 nm in the pcv to 437 nm in cmv to 435 nm in pm indicating that the environment of the hydrocarbon region of the lipids of H. halobium is more rigid than pcv and the pm lipids are 108 more restricted than the cmv lipids. AS in the pm had another peak at 417 nm. This emission maximum occurs at higher energies than the emission maximum of AS in rigid hydrocarbon medium (427 nm). The 417 nm emission probably originates from tightly packed AS molecules where the carboxyl group is further twisted out of plane of the anthracene ring due to the packing restrictions in the membrane such that the emission is more like anthracene than anthracene carboxylic acid in rigid hydrocarbon medium. The fluorescence wavelength maximum of the AS probe incorporated in the pm and cmv changed with temperature as shown in Fig. 6.7. To interpret such changes, one must point out that the fluorescence maximum of AS depends on the extent of excited state relaxation of the fluorescence moiety which in turn depends on the microscopic "viscosity" in the immediate vicinity of the fluorescence moiety. In other words, it depends on the free volume available for the motion of the anthracene ring. It is not surprising, therefore, to see that a phase transition temperature obtained from these measurements will not reflect the macroscopic viscosity of the membrane as measured by 5—NS or 12-NS ESR probes, but may correspond to a rotational degree of freedom in the fatty acid molecule: which may be probably the same in the pm and cmv. The tighter packing of the lipid molecules in pm due to the bR protein molecules will, however, limit the relaxation of the AS probe, compared to its relaxation in cmv giving rise to a smaller 109 .>Ec Away can Em Afiv Ca ma Mo mefixme Cowmmwso one mo accocComoc chapmnomEoP one 5.0 opswwm 0' o. nuap2 (amp/A)?- e -e A6 <70= max ...........7.2 where A is the bandwidth of the absorption band and is the wavelength of the monomer absorption maximum. 9(+ and 1_ are the wavelengths of the absorption maxima of the Split bands (Heyn, 1975). The absorption frequencies, dipole strengths and rotational strengths for various orientations of the exciton dimers have been calculated (Tinoco, 1963). This dimer model can be extended to trimers and infinite polymers. An excellent theoretical discussion about the excitonic interaction in a trimeric unit is given by Kriebel and Albrecht (1976). (e) Experimental absorption and circular dic hro mdata of pieached and spepwise reconstituted puppip membrane: We have obtained the absorption and CD Spectra of bleached and stepwise reconstituted pm to investigate the nature of chromOphore interactions of bRs in the pm. AS stated earlier, the color of pm is actually intense purple, similar to bright red color of dark adapted rhodopsin. In 122 the presence of visible light, the red color of rhodopsin changes to glossy yellow. This disappearance of the color of rhodopsin is referred to as "bleaching". BR cannot be bleached by mere exposure to visible light, it bleaches only with a harsh chemical treatment coupled with visible light. It can be reversibly bleached either by organic solvents such as ether, dimethyl sulfoxide (Hess & Oesterhelt, 1973) or by NH 1974). We achieved the stepwise bleaching of pm by suspending 2OH, pH 7.0 (Oesterhelt et al, it in 0.2 M NHZOH, pH 7.0, and exposing it to visible light between 500 and 700 nm, and monitoring the decrease in the absorbance at 560 nm with time (Fig. 7.1). The retinal Schiff base bond is broken by its reaction with NHZOH in the presence of light, forming a retinal oxime. Once this covalent bond iS broken, the interaction of retinal Schiff base with protein which is responsible for its absorbance diminishes, and there is an equivalent increase in the retinal oxime peak at 355 nm. Reconstitution was achieved by stepwise addition of 2 pl of 1 mM 13-cis or all-trans retinal to 3 ml of 10’5 M bR of washed, bleached pm. The purple color returned as a result of the reconstitution. After adding 13-ciS retinal, the bleached pm was incubated in dark for ten minutes (dark adapted). At low concentra- tions of retinal, only a small percent of the bRs were reconstituted, thus, most of the trimeric units had only one reconstituted bR (i. e. monomer). When the 123 .Psmpmsco ma E: mmm + xmcm E: omm mo zpflmsopr one ..::::::Em concmoan haaam va Em cmncmoap adamwppmm onnwacpnp ADV Em c>flpmc Amv .ceflp new: co>aomno 0.5 mg $0 $2 2 N.o ca mm A: muoH mo HE m v Em mo msflncmoan cmfismopm Mo muvcomm Ccfipmhcmn< .EQ ac mcflncmcan cmflsmopm mo appcomm :cflvmpomn< H.m ohswflm 124 .52 0mm 000 can own Om? Onw¢ 0.0M 00» q u a. 9 .. J w/HH../ / . . .. \ \ . (I. ./ \ \.\.\\“\ No l. c. \ \ \\\ a x. z/ a... \ .x w ’o o.‘oon\‘ \ \\ no... my I z .\ xx. ..... M I ./ .\ \\ \ an, .... .\ \ x R.,, . .. x . . cc \ \ w / \ .\ ..../ ....l. \\ o. I \. ... .. I.\ .. . . .. $0 a: ooooo 125 concentration of retinal was increased such that the retinal t0,bR ratio was 1.2:1, most of the trimeric units were fully reconstituted (i. e. trimers) (Becher & Ebery, 1976). The absorption maximum of the dark adapted monomer iS at 544 nm, that of the trimer is at 560 nm (Fig. 7.2). The retinal in organic solvents absorbs at 360 nm. If retinal would have been incorporated non-Specifically in the membrane rather than in the Specific binding Site (cavity), it would have absorbed at 355-360 nm instead of absorbing at 560 nm. During the reconstitution experiments, the intensity of absorption of retinal oxime peak remained unchanged until all the bRs were reconstituted. This observation indicates that no retinals were non-Specifi- cally attached to the other (lipid) regions in pm unless all the binding Sites in the bR were filled. In other words, the affinity of retinal towards the 'cavity' in the protein must be larger than towards the membrane lipids due to its amphipathic nature. The absorption spectra of the light adapted monomers and trimers were similarly obtained by adding various concentrations of all-trans retinal and exposing the sample to visible light between 500 and 700 nm for ten minutes after each addition. The absorption maximum of light adapted monomer occured at 551 nm, that of trimer occured at 568 nm, similar to that shown in Fig. 7.2. The Shift of the absorption to lower energies (approximate- ly 500 wavenumbers) compared with monomer absorption for the trimer absorption in both cases resulted from the 126 .HmCMPop mCMAPuaam Spas mp mo Ccfipapflpchcca cmflsmcpm mo mppcomm Ccflvmacmn< m.m chamfim mu: 00V 092 009 099 099 128 exciton interaction. AS discussed in chapter 4, three molecules of bR form a trimeric unit leading to P3 symmetry (each of the three molecules makes a Side of an equilateral triangle).From the electron microsc0py data, it seems that the retinal Schiff base situated in each of the three bR molecules also must be symmetrical around the P3 symmetry axis. However, the functional unit of symmetry of a tri- meric unit may be different than the structural unit (Kriebel & Albrecht, 1976). In the case of a monomer, the majority of the trimeric units contain only one bR molecule with a chromophore. Therefore, no exciton interaction is expected. The extent of exciton interaction for pm was shown to depend on the degree of reconstitution as shown in Fig. 7.2 as the increase in the red shift upon stepwise rec reconstitution. The head-to—tail packing of the chromOphoreS leads to a red Shift in the absorption maximum. The magni- tude of the red Shift depends upon (i) the angle between the plane of the membrane and the plane of the chromophore (ii) the dipole moment of the chromophore in the ground state and (iii) the distance between the chromophores. The distance between the retinal-Schiff base chromophores which have P3 symmetry can be calculated by the following equations: (i) For a degenerate transition A 2’3=1.75 B (6 ,¢)...7.3 (ii) For atotally symmetric stateA1=8.5 B (9, 4)) ....7.4 where B (9.4)) is a function of transition dipole moments. From the experimental red shifts, B (9 ,¢) can be 129 calculated. Further, R3 = )4; Ina. 4>>| / IB(e AMI ......7.5 where IMg is the ground state dipole moment of the chromo- phore, f(é, (p) is the dipole-dipole interaction which depends on the orientation of a chromOphore in a local co-ordinate system (Kriebel & Albrecht, 1976). From these calculations, an equilateral head to tail packing of three molecules of retinal-Schiff base at an angle of about 20° to the plane of the membrane can account for the CD data and would explain the observed absorption red Shift. Our CD Spectra of stepwise reconstitution agreed with the absorption data (Fig. 7.3). The documented red shift in the CD spectra of the positive lobe and the increase in the intensity of both, especially the negative lobe, as a function of the degree of reconstitution was confirmed (Heyn et al, 1975),(Becher & Ebery, 1976),(Bauer et al, 1976),(Ebery, 1977). Furthermore, although the CD Spectra showed a positive peak at 355 nm (Fig. 7.3), no interpreta- tion has yet been given. Our experiments Show that the magnitude of this 355 nm peak is inversely proportional to the degree of reconstitution. At 100% reconstitution, the 355 nm peak had approximately 10% intensity compared with the peak of the bleached pm. On the other hand, the 355 nm peak of the absorption Spectra of the stepwise reconstitu- tion remained unchanged upto 100% reconstitution. As 130 .opacfl£\ s: 0H was comma .mcsoccm : mm; paumCOC cefipcnp .80 Ace mcoawccfiaawe N was camcm no one .Hs\ws cow was as one Mo Ccflpmapaoccco one .oppo>ac amasmcmPooa .Eo occ m Ga 0.5 mm mane S Ho.osw conceamSm was am one .com: mm: ocflpmcwmwcoe no omumm cwmwpscwom Abommm a new: mn>p\amo Hcccs oom