MSU

LIBRARIES

\—

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

 

PHOTOELECTROSPECTROMETRY OF BILAYER LIPID MEMBRANES (BLM) AND
PARTIAL RECONSTITUTION OF THE PHOTOSYSTEM 1 REACTION CENTER
OF PLANTS IN ARTIFICIAL MODEL MEMBRANES

By

Jose R. Lopez Santiago

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Physiology and Biophysics

1982

60.00%

ABSTRACT

PHOTOELECTROSPECTROMETRY OF BILAYER LIPID MEMBRANES (BLM)
AND PARTIAL RECONSTITUTION OF THE PHOTOSYSTEM 1 REACTION CENTER OF
PLANTS IN ARTIFICIAL MODEL MEMBRANES

By
Jose R. Lopez Santiago

BLMs modified with different substances were investigated for
photoelectric effects when illuminated with uv and visible light. One
of the systems studied consisted of membranes formed with extracts from
spinach chloroplasts. Another system investigated consisted of an
otherwise non-photoactive membrane to one side of which liposomes
containing purple membrane (PM) fragments from H. halobium were added.
Both of these BLMs exhibited photopotentials when illuminated with uv
and visible light. In both cases the photovoltaic action spectrum
followed the absorption spectrum of the modifier. In the system with
PM, through chemical modification of the PM, the uv-elicited
photoresponse was found to be due to energy transfer from the aromatic
amino acids to the chromophore and subsequent proton translocation by
the chromophore. One other system investigated consisted of BLMs

containing chymotrypsin adsorbed to one side. This BLM also exhibited

photovoltages when irradiated with uv light whose action spectrum
follows the absorption spectrum of chymotrypsin. The initial charge
separation in this case is attributed to ionization of the aromatic
amino acids by the uv radiation. The results obtained in this set of
experiments suggest the possibility of using photoelectric spectroscopy
measurements to study energy transfer and interaction between membrane
components in BLMs.

In another set of experiments, the partial reconstitution of the
P51 reaction center (RC) from plants in BLMs was accomplished by two
different methods. The first method consisted of the association of
vesicles containing thylakoid membrane fragments (TMFs) with the BLM.
The second method involved the incorporation of the TMF in the BLM
through fusion of vesicles with the BLM. In both cases photoelectric
effects were observed which required the addition of carriers of
reducing equivalents. The action spectrum of the photoelectric
response followed the action spectrum of the P51 RC. This set of
experiments gave support to the idea that charge separation in the PSI
reaction centre results in the generation of a potential difference
across the thylakoid membrane. The results also support the view that
the RC spans the thylakoid membrane. Moreover they show the usefulness

of the planar artificial BLM for reconstitution studies.

To
my parents,
my sister Nilda,
my brothers Pedro and Carlos and

especially my wife Doris

ii

ACKNOWLEDGEMENTS

I wish to express my appreciation to my advisor, Professor H.T.
Tien, for suggesting the topic of this dissertation and for his
guidance, help and advise during the course of this study.

I also want to express my deepest thanks to my friends and
colleagues Qi-Yi Liu, Qing-Yu Hu, Denise Mazorow, Dr. John Higgins,
Kevin O'Boyle, Dr. Joseph Zon and Dr. N.B. Joshi for their invaluable
help, suggestions and friendship during the course of my studies at
Michigan State University. I want to give my thanks also to Drs. E.
McGroarty and A. Haug for allowing me to use some of their facilities
and to Dr. E. Eisenstein for agreeing to continue to serve on my
dissertation committee. Thanks go to all those which in one way or
another contributed to the successful completion of this project.

Very special thanks to my wife Doris for her support, patience and
encouragement and also for typing the rough draft of the dissertation.

Finally, I want to thank the University of Puerto Rico for
allowing me the opportunity and time necessary for the completion of my

studies.

This work was supported by a National Institute of Health Grant

(GM14971).

TABLE OF CONTENTS

Chapter 1 - Introduction................ ..... . ................. .....1

The BLM Model Membrane
(a) General.............................................6
(b) Photoelectric effects on BLMs

containing chlorOphyll pigments.....................8
Review on Reconstitution in Planar Membranes
(a) Introduction.......................................10
(b) Methods and techniques for the

reassembly of proteins and membrane

fragments in planar membranes......................12
(c) Biological membrane components

reconstituted in planar membranes..................18
The Photosynthetic Apparatus of Plants
(a) The chloroplasts........ ......... . ....... . ..... ....25
(b) The thylakoid membrane.............................26
(c) Charge separation and transport

of electrons and protons in

plant photosynthesis......... ..... .................29
(d) Electric potential across the

thylakoid membrane and ATP synthesis...............34

iv

Chapter 3 - Experimental Methods and Materials

Materials
(a) Chemicals..........................................40
(b) Composition of solution used

for forming BLMs .......... . .................... ....40
(c) Purple membrane....................................41
(d) Thylakoid membrane fragments.......................41
(e) Experimental set-up for electrical

measurements and formation of BLM..................41
Methods
(a) Formation of BLM................ ................ ...43

(b) Culture conditions for H. halobium.................47

 

(c) Harvesting.........................................47
(d) Isolation of purple membrane.......................47
(e) Chemical modification and/or

bleaching of PM....................................48
(f) Isolation of chloroplasts from

spinach leaves.......................... ........... 57
(9) Isolation of thylakoid

membrane fragments.................................57
(h) Preparation of liposomes or

vesicles containing PM or TMFs.....................59
(i) Fusion of vesicles with BLMs.......................59
(j) Incorporation of proteins

into BLMs................. ....... ..................62

(k) Electrical measurements. ..... ......................62

Page
Chapter 4 - Photoelectrospectrometry of BLM

Introduction........................ ................... 64
Photoelectric Spectroscopy of Chemically

Modified PM Reconstituted on BLMs

(a) Background......................... ................ 66
(b) PM modified with N-bromosuccinimide................67
(c) Bleaching with hydroxylamine .......... . ............ 71
(d) Bleaching with NaBH4...............................74
(e) Summary............................................77

Chapter 5 - Partial Reconstitution of the P31 Reaction

Center of Plants in BLM Model Membranes

Introduction

(a) General............................................81
(b) Background ...... . ........... .................. ..... 82
Results

(a) Association of TMF-vesicles

with planar BLMs................ ................... 82
(b) Kinetics of the association process................83
(c) Action Spectrum of the photoresponse ............... 92
(d) Open-circuit photovoltage..........................92
(e) Short-circuit photocurrent........................104
(f) Incorporation of RCs from thylakoid

membranes into planar BLMs........................118
Discussion
(a) Association of TMF-vesicles

with BLMs ........................................ 122

vi

(b) The photoelectric response.... ...............

(c) Incorporation of RCs from thylakoid

membranes into planar BLMs....................

(d) Proposed model for the photoelectric

response of the two TMF-BLM-vesicle

systems studied..... ..... . ............ . ......

BibliographyO0.00...OOOOOOOOOOOOOOOOOOOO0.0.000...

Appendix A000. ........... O... 000000 .0... 000000000
Appendix B .......................................
Appendix COOOOOOOOOOOOOOOOOOOO. 00000 O 000000000000

vii

Page

.....123

000.129

..... 131

0000135

..... 145

..... 157

LIST OF FIGURES

Page
Figure 2.1 Z-scheme of plant photosynthesis........ ............... 32

Figure 3.1 Block diagram of the experimental

arrangement used in the BLM studies....................45
Figure 3.2 Schematic of the circuit used to monitor

the BLM conductance....................................46
Figure-3.3 Absorption Spectrum of purple membrane from

bacteria grown in the laboratory............... ........ 50
Figure 3.4 Absorption spectra of purple membrane

modified with NBS......................................52
Figure 3.5 Absorption spectra of purple membrane

bleached with NH20H4 in the presence

of light................................ ...... . ........ 54
Figure 3.6 Absorption spectra of purple membrane

bleached with NaBH4 in the presence

of light.................................. ..... ........56
Figure 3.7 Electron micrographs of the vesicles

used in these studies (34,560X) -

A) PM-VESiCIeSO B) TMF-veSTCIESoooooooooooo00000000000061

viii

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 5.1

Figure 5.2

Photovoltaic action spectra from BLM

fused with vesicles containing

purple membrane modified with NBS...... ................ 70
Photovoltaic action Spectra from BLM

fused with vesicles containing purple

membrane bleached with NHZOH... ........................ 73
Photovoltaic action spectra from BLM fused

with vesicles containing purple membrane

bleached with NaBH4 and light..........................76
A) Change in the height of the 280 nm

peak relative to that of the 560 nm peak

as a function of NBS concentration. The

change is expressed as the ratio of

photovoltage at 280 nm. B) Change in the

heights of the 280 and 360 nm peaks relative

to the 560 nm peak as a function of

illumination time in the presence of NaBH4.

The changes are expressed as ratios of

photovoltages at given wavelengths ..................... 79

Open-circuit photovoltage generation by

TMF-vesicles added to one side of a BLM................85
Time course of the photovoltage development

for different amounts of vesicles added

to the aqueous phase ................................... 88

ix

Figure 5.3

Figure 5.4

Figure 5.5

Figure 5.6

Figure 5.7

Figure 5.8

Figure 5.9

Figure 5.10

Figure 5.11

Variation of the photovoltage as a

function of the aqueous phase

vesicle concentration......... ...... .......... ......... 89
Time course of the photovoltage

development for vesicles containing

different amounts of Chl a.............................91
Variation of the photovoltage as a

function of the amount of Chl a in

the vesicles.................................... ....... 93
Action spectra of the P31 RC (from Ried,

1972) and photocurrent action spectrum

of the photoelectric response for the

TMF-vesicles associated with the BLM...................95
Effect of C006, o-phenanthroline, CCCP

and of external shunt resistance on the

open-circuit photovoltage....... .......... ..... ........ 97
Time course of the open-circuit

photovoltage in the presence of Vit K3.................99
Variation of the open-circuit

photovoltage with light intensity.....................101
Time course of the open-circuit photovoltage

in response to illumination with an 8 us

in duration flash of light............................102
Development of the slow component of the

photovoltage in the presence of 0.5 mM PMS ...... ......103

Page

Figure 5.12 Time course of the short-circuit photocurrent

for TMF-vesicles associated with a BLM. .............. .105
Figure 5.13 Diagram representing the suspected

configuration of the vesicle-BLM system

after association of the two

membraneous structures............. ................... 107
Figure 5.14 Equivalent circuit diagram to represent

the vescile-BLM associated system as

shown in Figure 5.13.................... .............. 108
Figure 5.15 Relationship between short-circuit

current and intensity of illumination. ................ 112
Figure 5.16 Variation of the measured time constant

for the short-circuit current (1') with

light intensity.......................................115
Figure 5.17 Variation of short-circuit photocurrent

with CCCP concentration............ .............. .....117
Figure 5.18 Increase in the BLM conductance in the

presence of hyperosmotic TMF-vesicles .............. ...119
Figure 5.19 Open-circuit photovoltage for a BLM

with incorporated TMFs........... ........... ..........121
Figure 5.20 Schematic of the electrical circuit

which approximately represents the

system consisting of vesicles associated

WithaBLMCOOOOOOOOOOO OOOOOOOOOOOOOO 00.0.0000000000000128

xi

Figure 5.21 Diagram of the pr0posed model to
explain the possible sequence of
events taking place across the thylakoid
membrane responsible for the observed

photoelectric responSEOOOOOOOOOO..OOOOOOOOOOOOOOOOOOO.133

xii

CHAPTER 1

INTRODUCTION

Membranes comprise a large part of cell structure. From a
functional point of view, membranes are known to be involved in many of
the essential processes taking place in living organisms. These
include: transport phenomena, energy transduction, excitability, exo-
and endocytosis, DNA anchoring, protein synthesis, cell-cell
recognition, synaptic transmission drug and/or hormone interaction and
many others. The diversity of the processes reflects the structural
differences and variations in the membrane components. In most cases,
it has been possible to associate certain membrane components, e.g.
lipids, proteins, lipoprotein complexes, pigments and other molecules
with the different membrane functions. In the hope of understanding
how biological systems work, it seems therefore necessary to obtain
information on the molecular aspects of the processes, on the
physico-chemical properties of the components involved, on the
interaction between them and also on structure-function relationships.

In many cases, however, the direct investigation into the function
of biological membranes is hampered by their structural complexity.
This has driven researchers into developing membrane models which mimic
certain aspects of biological membranes. This approach has the

advantage that in model systems, in general, some of the properties of.

1

the physiological system are abstracted from the many others which make
the study of the whole system more difficult. Moreover, model membrane
systems provide a more controlled environment and help gain access to
certain experimental parameters which are otherwise inaccesible in the
natural system. One important membrane component which has been
extensively investigated in several different model systems is the
chlorophyll molecule (for a review see Seely, 1977).

One such model system which has been extensively used is the
so-called bilayer (or bimolecular) lipid membrane (BLM) developed by
Mueller, Rudin, Tien and Hescott in the early 1960's (Mueller, Rudin,
Tien and Hescott, 1962; 1963). One of the main advantages of this
artificial membrane system is that its planar configuration allows
direct characterization of electrical events taking place across it.

One area of research using the BLM model system involves the
reassembly of components of biological membranes into the model system
and the study of its properties and characteristics (for a
comprehensive review see Tien, 1974). Following the same approach,
Tien and co-workers formed BLMS with a solution containing pigments and
lipids extracted from chloroplast lamellae and discovered that such
membranes were capable of generating photoelectric phenomena (Tien,
1968; Tien and Verma, 1970). The photoelectric effects have been
attributed to electronic processes taking place inside the membrane
coupled with redox reactions taking place at the interfaces (Tien,
1972; 1979). Since then many other photoactive substances have been
found to generate photoelectric effects in BLMS, including dyes (Tien,
1972; Ullrich and Kuhn, 1972; Huebner, 1978), carotenoids (Kobamoto and

Tien, 1971), light-sensitive inorganic ions (Pant and Rosenberg, 1971),

purple membrane from H. halobium (Dancshazy and Karvaly, 1976; Herrmann

 

and Rayfield, 1978; Higgins, Lapez and Tien, 1979; see also Appendix
B), porphyrin derivatives (Hong and Mauzerall, 1972), porphyrin
complexes (Joshi, Lopez, Tien, Wang and Liu, 1982; see also Appendix
C), rhodopsin (Liu and Tien, 1982) and reaction centers from
photosynthetic bacteria (Packham, Packham, Mueller, Tiede and Dutton,
1980).

One apsect of the study of photoelectric phenomena in photoactive
BLMS is that of photoelectrospectrometry. The term
photoelectrospectrometry is used here to describe the measurement of
the action spectrum of the photoelectric effect, which Should resemble,
in general, the absorption spectrum of the photoactive Species in the
model system. From this follows the most immediate and common use of
the action Spectrum that is as an indicator of the species responsible
for the initial light-absorbing step in the photoelectric process (see
Appendix C). However, Since the photoactive species is present in a
membrane environment which is usually different from bulk solutions,
the action Spectrum may also exhibit some differences from the
absorption spectrum which reflect the influence from that environment.
It also happens that direct measurement of the absorption spectrum of
the species in the membrane phase is very difficult due to the
ultrathinness (m 100 A) of the BLM (Cherry, Hsu and Chapman, 1971). On
the other hand, the action spectrum of the photoeffects, due to its
nature, is in reality an excitation type of Spectrum and not an
absorption spectrun (as in the case, for example of fluorescence
excitation spectrum). Therefore it may also reflect differences from

the absorption spectrum and should reflect characteristics of the

photoelectric effect itself. This is exemplified by studies on BLMS
containing chlorophyll pigments (Van and Tien, 1970) and purple

membrane from H. halobium (Karvaly and Dancshazy, 1977). In both of

 

these studies, differences between the action spectrum of the
photoeffects and the absorption spectrum of the photoactive Species
were observed. The differences were explained in terms of the
properties and characteristics of the photoactive Species and the
nature of the photoeffects.

This research centers around two aspects of the reassembly of
membrane components into BLMS, mainly photoelectrospectrometry and
reconstitution. In the first case, the main objective of the
research is to perform action Spectrum measurements of BLMS containing
several photoactive Species including proteins. The action spectrum
will be measured in the visible as well as in the uv region which is
where most proteins absorb. It is hoped that the Spectra may give
information on the interaction and/or energy transfer between the
membrane components.

The other main objective of this research is to go one step ahead
in the reassembly of components of the photosynthetic apparatus of
plants and try to reconstitute, at least partially, the photosystem 1
(PS1) reaction center of plants in a BLM model system. The word
reconstitution is meant to describe the reassemly of the component in
‘such a way that at least some of its functional activity is retained.
Under such circumstances it may be possible to obtain direct evidence
that a potential difference develops across the thylakoid membrane of
chloroplasts as a result of illumination. Reconstitution experiments

of this type have been successful in the case of isolated reaction

centers from photosynthetic bacteria (Barsky gt_al., 1976; Packham gt
$1., 1980). However, when working with isolated PS1 reaction centers
from plants the results have not been as successful (Barsky gt 11.,
1976; Lopez, unpublished observations). For this reason it was decided
to try to reconstitute not the isolated reaction centers but whole
membrane fragments obtained from thylakoid lamellae.

The results of the experiments on photoelectrospectrometry appear
in Chapter 4 and also in Appendix A. The results of the experiments
on the partial reconstitution of the thylakoid membrane fragments are
presented in Chapter 5. Also, because of the wide range of subjects
dealt with in this research the literature review has been divided into
three parts: the BLM model system, the photosynthetic apparatus of
plants, and reconstitution of proteins from biological membranes into
planar model membrane systems. This literature review is presented in

Chapter 2.

CHAPTER 2

LITERATURE REVIEW

THE BLM MODEL MEMBRANE

(a) General:

Biological membranes are complex and highly variable structures.
The variability is reflected in both function and composition. At the
same time, however, there are also some properties and characteristics
of biological membranes which are common to almost all biological
membranes found in living organisms. On the one hand biological
membranes are usually composed of two classes of macromolecules,
proteins and lipids. Also, stained electron micrographs of membranes
from different sources all exhibit the same pattern, two dark electron
dense regions sandwiching a not so dense light region. From electron
microscope studies the thickness of most biological membranes has been
found to range between 40 and 110 A. It has been proposed that all
membranes have the same basic structure and a large number of models
have been proposed over the years since the original proposal of the
bimolecular leaflet model of Carter and Grendel (1925), followed later
by the "unit membrane" hypothesis of Robertson (1959). Up to now, no
single model can explain all of the data obtained on the structure of
biological membranes, however a large number of observations can be

explained in terms of the model of Singer and Nicholson (1972). The

6

model proposes a fluid lipid bilayer interdigitated by proteins.

As mentioned previously, due to the complexity of biological
membranes, several investigators decided to develop and study membrane
models. Of the several model systems which have been developed,
the liposome, introduced by Bangham (1963) and the BLM developed by
Rudin and co-workers (Mueller gt 31., 1962) have received the most
attention. Both of these artificial membrane systems consist of an
ultrathin (m100 A in thickness) lipid bilayer matrix separating two
aqueous phases. The popularity of both these models lies mainly in the
ease of their preparation and handling, the way they mimic biological
membranes under certain circumstances and their Significant role in
membrane reconstitution studies. Several detailed reviews on both of
these model systems have been published over the years (see Shamoo and
Trivol (1980), Tien (1974) and Tien (1981) for reviews on BLMS and see
Bangham, Hill and Miller (1974), Szoka and Papahdjopolous (1980) and
Tyrell, Heath, Colley and Ryman (1976) for reviews on liposomes).

Of the two model systems, the BLM has proven especially useful in
studies involving electrical characterization, diffusion and
permeability since its planar configuration allows easy access to both
sides of the bilayer.

Some of the areas of study involving BLMS include: formation
techniques and materials (Takagi, Azuma and Koshimoto, 1965; Tien,
1974; Vodyanoy, and Murphy, 1982; Naldbillig and Szabo, 1979)
mechanical, physical and chemical properties of BLMS, e.g. thickness
(Tien, 1974), stability (Chizmadzhev, Abidor, Postushev and Arakelya,
1979), bifacial tension (Tien, 1974), fluorescence spectroscopy

(Alamuti and Lauger, 1970, electrical breakdown (Benz, Janko and

Lauger, 1979), diffusion and permeability (Fettiplace and Haydon;
Orbach and Finkelstein, 1980) and ion transport (Andersen, 1978;
Jordan, 1980); the effect of antibiotics on membrane permeability
(Andreoli and Tosteson, 1971; Muller and Rudin, 1967); the interactions
of toxins with BLMS (Donovan, Simon Draper and Montal, 1981); the
interaction of liposomes with BLMS (Duzgunes and Ohki, 1981;
Zimmerberg, Cohen and Finkelstein, 1980); reconstitution of membrane
components on planar membranes (Montal, Darszon and Schindler, 1981;
Shamoo and Trivol, 1980).

(b) Photoelectric effects on BLMS containing chlorophyll

 

pigments:

Photoelectric phenomena on BLMS was first observed by Tien (1968)
on artificial membranes formed with an extract obtained from
chloroplast lamellae (Chl-BLM). Since then, photoelectric effects for
several other modified BLM systems have also been observed, including
BLMS containing; carotenoids (Kombamoto gt 31., 1971), purple membrane

from H. halobium (Dancshazy gt al., 1976; Herrman gt_gl,, 1978; Higgins

 

.gt‘g1., 1979), porphyrin derivatives (Hong gt 31., 1972), prophyrin
complexes (Joshi gt_al., 1982; Wang, Tien, Lopez, Liu, Joshi and Hu,
1982), rhodopsin (Liu gt 31., 1981; Montal, 1979) and
bacteriochlorophyll reaction centers (Packham gt_g1,, 1980; Schonfeld,
Montal and Feher, 1979). Photoelectric effects have also been observed
when certain modifiers (so-called extrinsic) are added to the bathing
solution at either side of a BLM, including: dyes (Huebner, 1978; Tien,
1972; Ullrich gt 21., 1972) and light-sensitive inorganic ions (Pant
and Rosenberg, 1971). Several reviews on this subject have been

published and the reader is referred to these for more details,

including Tien (1974) and Tien (1979). (Two of the previous systems,
namely, purple membrane-BLM and bacteriochlorophyll reaction centers -
BLM will be discussed in more detail in the next section).

The observation of photoelectric effects on BLMS suggest the
generation of light-induced charge carriers in the membrane.
Originally, the photoelectric effect from a Chl-BLM was found to be
very small since the system was set up to be symmetric (Tien, 1968).
Later the photoresponse was found to increase dramatically under
assymetric conditions, such as: externally applied electric field, the
presence of electron acceptors and/or donors on opposite Sides of the
BLM and also a pH gradient across the BLM (Tien, 1974). The
photoelectric effect consisted of a photovoltaic and a photoconductive
effect. The action spectrum of the photovoltaic effect identified the
chlorophyll pigments as the photoactive species responsible for the
photoeffects. The photovoltage was found to vary linearly with an
externally applied potential difference. This is attributed to the
large magnitude of the electric field (up to 105 V/cm) present across
the membrane which can cause the oppositely charged species, generated
either directly or indirectly by light, to move to opposite sides of
the BLM depending on the direction of the electric field. Such a
tremendous electric field is possible due to the small thickness (W 100
A) of the BLM and its ability to withstand potential differences of 100
mV or more. The magnitude of the open-circuit photovoltage was found
to increase with the logarithm of the incident light intensity.
Moreover, when examined under continuous illumination the photovoltage
was found to consist of two components, with one much faster than the

other. By illuminating the BLM with a short duration (8 us) light

10

flash, the rise time of the first component was found to be faster than
8 us (Heubner and Tien, 1972).

AS mentioned previously, it was found that addition of certain
oxidizing agents (electron acceptors) to one Side of the BLM, enhanced
dramatically the photoresponse (Tien and Verma, 1970). At the same
time, addition of electron donors (reducing agents) to the opposite
side of the membrane further enhanced the photovoltage and/or
photocurrent (Tien, 1974). Both of these facts taken together have
been interpreted to indicate that electron transfer reactions are
taking place at each interface (Ilani and Berns, 1972; Tien, 1968;
TriSSl and Lauger, 1972). The question is then 'What is the nature of
the charge carriers inside the bilayer?' Two mechanisms have been
proposed to answer this question. In one case the migrating species
are said to be radicals (Trissl and Lauger, 1972). In the other
mechanism it is proposed that electrons and holes are the charge
carriers, suggesting that the membranes are capable of electronic
conduction (Ilani gt_al., 1972; Tien, 1968). It has been suggested
that carotenoids, which are present in the BLM, provide a pathway of
low enough resistance to electron movement across the "bulk" membrane
(Mangel, Berns and Ilani, 1975). For more details on the mechanism for
electronic conduction by electrons and holes across the Chl-BLM the

reader is referred to a recent review by Tien (1979) on this subject.

REVIEW ON RECONSTITUTION IN PLANAR MEMBRANES

(a) Introduction:

 

Recently, an increased nunber of reports on reconstitution of

certain membrane components on planar bilayer (model) membranes have

11

appeared in the literature (for a review see Montal, gt al., 1981).
This increase in the number of reports follows some new methods and
techniques used to incorporate membrane components, including proteins
and lipoprotein complexes, into planar bilayer membranes (see Montal gt
.11., 1981). This is tied in with some recent progress in membrane
research which makes it possible to associate complex biological
processes with membrane components, including protein lipids, pigments
and others, and with technical advances in the biochemical
characterization and purification of those components.

The main features of the reconstitution approach are that it
"allows, in principle, the dissection of a complex phenomenon into its
components with the subsequent gain of experimental control over
parameters that are otherwise inaccesible or highly varible in the
natural membrane". Also "it establishes general principles for the
reassembly of biologically active membranes providing insight into the
interplay between the membrane environment and the functional unit“.

Model membranes have proved very valuable in membrane
reconstitution studies in terms of providing a more controlled
environment. The most widely used model membranes are the liposome or
vesicle of spherical configuration and the planar bilayer lipid
membrane, first introduced by Mueller gt al. in 1962. Liposomes have
been used widely especially in spectroscopic studies and have the
advantage of being easy to handle, relatively stable and that proteins
are relatively easy to be incorporated in them. Planar membranes, on
the other hand, although more unstable have the advantage that they
allow easy access to both of its Sides and therefore provide a direct

way for the characterization of electrical events taking place

12

membranes. Unfortunately, the functional reassembly of proteins in
planar membranes is difficult and has therefore advanced more Slowly.
As mentioned previously however, recent basic conceptual and
methodological advances in protein reconstitution in planar membranes
are providing new hope and perspectives in this approach.

In this review section mostly reconstitution studies involving
planar membranes will be discussed although in some cases and when
relevant, work done with model membranes of special configurations will
also be mentioned.

(b) Methods and techniques for the reassemb1y_of proteins and

 

membrane fragments in planar bilayers:

 

At present time there are two main approaches used to incorporate
proteins into planar membranes. One approach involves the formation of
planar bilayers in such a way that the proteins become incorporated at
the moment of formation. In the other approach the proteins are added
into the aqueous phase of an already preformed planar membrane with the
protein adsorbing into the membrane and eventually becoming
incorporated into it. In both cases the planar bilayers can be formed
either as a bimolecular (black) lipid membrane (Mueller gt_al., 1962)
or assembled from monolayers (Takagi gt al., 1965).

In the simultaneous approach the material to be incorporated has
to be either suspended in a lipid organic solvent phase (Montal, 1974)
or incorporated in a vesicle suspension (Schindler and Quast, 1980).
The rationale for the transfer of proteins in active form into organic
solvents is based on the fact that membrane proteins usually have
hydrophobic and hydrophilic regions. It can be assumed that the

hydrophobic parts will be at least partly soluble in the apolar

13

solvents while the hydrophilic regions will not. A procedure had been
established by Das and Crane in 1964 in order to extract complexes of
cytochrome c and acidic and neutral lipids in isooctane. They found
that the partition of the complexes into the apolar solvents was
enhanced by the charge neutralization between the positive charges of
cytochrome c and the acidic phospholipids. Then Gitler and Montal
(1972) found that the partition of these complexes in apolar solvents
could be enhanced by adding cations like Ca2+ and Mg2+ and by

lowering the pH. Following these results several groups have succeeded
in transfering certain protein and proteolipid complexes into alkanes.
These include cytochrome oxidase lipid complex in hexane (Montal, 1974)
and octane (Chien and Mueller, 1976), rhodopsin in hexane (Darszon,
Philipp, Zarco and Montal, 1978) and octane Liu gt 11., 1982) and
reaction centers from photosynthetic bacteria in hexane (Schonfeld, gt
31., 1979, Kendall-Tobias and Crofts, 1979) and octane (Packham gt 21.,
1980).

Two methods are at present generally used to transfer the proteins
into the apolar solvent. In one case the proteins are first
solubilized by detergent, purified and delipidated if needed. Then the
protein can be added directly to solvent containing lipids or
recombined with lipids to form vesicles which are then extracted with
the organic solvents. This allows for some control over the protein
and its reconstituted lipid environment. In the other methods, the
protein is transfered directly into the non-aqueous phase from the
biological membrane. This is accomplished by mixing the biological
membrane with the organic solvent and sonicating followed by addition

of cations and separation of the two phases by centrifugation.

14

One other interesting technique, which follows the Simultaneous
approach, was recently developed by Schindler (1980) and it is based on
the results of an experiment by Verger and Pattus (1976) (see also
Pattus, Desmuelle and Verger, 1978; Pattus, Piorant, Lardunski,
Desmuelle and Verger, 1978 and Schindler, 1979) who found that
monolayers can form Spontaneously at air water interfaces containing
liposomes, or vesicle suspensions derived from biological membranes.
Using this result and following the method of Takagi gt_gl,, (1965),
Schindler (1980) was able to form planar bilayer membranes. The
membranes were found to be stable if the liposomes or vesicles used for
spreading the monolayers were > 50 nm in diameter. It was found that
if vesicles obtained from biological membranes containing proteins were
used to Spread the monolayers, the planar bilayers obtained from them
would also contain the proteins (Schindler gt 31., 1980). Using this

technique a functional acetyl choline receptor from Torpedo marmorata

 

was reconstituted in planar membranes (Schindler gt_gl,, 1980). The
technique also worked when the acetyl choline receptor was isolated,
purified, and incorporated into liposomes which were spread to form
monolayers and eventually planar bilayers (Nelson, Anholt, Lindstrom
and Montal, 1980). The results obtained in this study were similar to
the ones obtained with native bio-membrane vesicles by Schindler gt
31., (1980).

The second approach followed in planar membrane reconstitution
studies known as the sequential approach is the one which has been most
widely used. In this approach the material to be incorporated can be
added to the aqueous phase in purified form, as a detergent or as a

vesicle suspension. This latter technique has been developed into what

15

is referred to as fusion of vesicles with planar bilayers. The
disadvantage of using detergent extracts is that detergents usually
have a deleterious effect on the planar membrane.

The technique of membrane fusion for reconstitution studies was
originated by Skulachev and co-workers in studies with cytochrome
oxidase, H+-ATPase and bacteriorhodopsin (Drachev, Jasaitis, Koulen,
Kondrashin, Liberman, Hayrecek, Ostroumov, Semenov and Skulachev,
1974). It was found, however that total fusion of the vesicles was not
taking place since the measurements were reproduced on thick planar
membranes. The results were attributed to adsorption or partial fusion
where the vesicle closed structure is maintained after attachment of
the two structures (Drachev, Frolov, Kaulen, Kondrashin, Samuilov,
Semenov and Skulachev, 1976; Herrmann gt_gl,, 1978).

More successful reconstitution using fusion of vesicles with
planar membranes was accomplished by Miller and Racker (1976) in order
to study the permeability properties of the sarcoplasmic reticulum
(SR). AS a result of these studies, an experimental protocol was
devised which consisted of making SR vesicles interact with planar
bilayers in the presence of negatively-charged lipids, calcium and an
osmotic gradient across the vesicular membrane leading to the swelling
of the vesicles (Miller, Arvan, Telford and Racker, 1976). The
criterium for fusion was the observation of an increase in the bilayer
conductance in discrete steps which were associated with fusion events.
The conductance increases were composed of a sequence of spike-like
events of varying size. It was suggested that each spike represents
the fusion of a single SR vesicle with the planar bilayers Miller‘s;

gl., 1976; Miller, 1978) resulting in the insertion of package of

16

potassium channels. Incorporation of proteins for reconstitution
studies using fusion of liposomes or vesicles with planar bilayer
membranes has also been reported by other groups, including Repke,
Berczi and Matties (1980), Hanke, Eihl and Beheim (1981) and Latorre,
Vergara and Hidalgo (1982). The results of some of these experiments
although all of them positive, sometimes contradict each other as well
as previous results. In the case of Tepke gt 31., (1980) no
requirement for osmotic gradient nor enhancement of fusion events using
Ca2+ were observed. The planar membranes used in this case were of
the black lipid membrane type. Fusion was said to depend on the type
of lipids present in both membrane structures. Hanke gt 21., (1981) on
the other hand, using planar membranes formed according to the method
of Takagi gt 31., (1965) found that while the presence of Ca2+ ions

is not a prerequisite for fusion, application of an osmotic gradient
appears to be necessary for high fusion rates. It was also found that
maximum fusion rates occur at the calorimetric phase transition
temperature and that the addition of cholesterol to the membrane lipid
stops fusion. Latorre gt_§1,, (1982) followed the protocol of Miller
gt_§1,, (1976) in their studies. In all previous cases the criteria
for successful membrane fusion was the observation of sharp increases
in the membrane conduction indicative that the channel forming proteins
had been incorporated in the membrane.

Fusion of liposomes or vesicles with planar membranes has also
been observed by other groups interested in studying the process of
exocytosis (Zimmerberg, Cohen and Finkelstein, 1980; Duzgunes and Ohki,
1981). In both of these studies the divalent ion Ca2+ was found to

be a requirement for fusion to cocur. Duzgunes gt_gl., (1981) used

17

gramicidin A present in the liposomes as the assay for fusion with
appearance of conductance channels in the planar membrane as the
criterium that fusion had taken place. Zimmerberg gt al., (1981) and
Cohen gt g1., (1981) found that an osmotic gradient across the planar
membrane was a requirement for fusion to take place. In these studies
the authors used multiamellar vesicles containing a voltage-dependent
anion channel (VDAC) from mitochondria incorporated in n number
lamellae and the fluorescent dye carboxy-fluorescein incorporated in
the aqueous space between the lamellae. The two criteria for fusion
included: 1) the incorporation of the VDAC protein in the planar
membrane as assayed by monitoring changes in conductance and 2) the
transfer of an n-1 lamellar liposome across the planar membrane as
assayed by detecting the fluorescent liposomal particle in the opposite
Side to the one in which it was originally located. Both of these
results were observed in these experiments and they were explained in
terms of the n lamellar liposome (containing the VDAC) fusing its outer
membrane with the planar membrane and the n-1 lamellar particle moving
to the other Side of the planar membrane. The discharge of vesicular
contents across the planar membrane represents the most convincing
evidence of vesicle-membrane fusion hitherto obtained.

It Should be mentioned that in the studies mentioned above true
fusion between the two membranes has not been unequivocally shown
except perhaps in the studies by Zimmerberg gt al., (1980). The reason
for this being that, as mentioned earlier, in most cases the assay for
fusion involves the incorporation into the planar membrane of a
membrane associated marker from the vesicular membranes. Some authors

have not ruled out (Cohen and Moronne, 1976) that the incorporation of

18

such markers can occur via its transfer by a process other than fusion.
However, for reconstitution studies the important thing is that the
material intended for reconstitution becomes incorporated in a
functional way independent of the process used. It is also important
to indicate that all the data obtained from biological membrane systems
reconstituted using different techniques and different preparations
will provide information that help in trying to resolve the question of
how the particular system works.

(c) Biological membrane components reconstituted in planar model

 

membranes:

One biological membrane component which has been reconstituted in
planar membranes by various research groups iS the protein
bacteriorhodopsin which is present in the purple membrane patches of H;
halobium. The first report was published by Drachev gt 51., in 1974.
In this study purple membrane containing bacteriorhodopsin was mixed
with phospholipids and incorporated first into liposomes which were
then added to one Side of a thick (not bilayer) planar phoSpholipid
membrane in the presence of Ca2+ ions. After waiting for some time
the system developed photopotentials when illuminated. The action
spectrum of the photopotentials followed the absorption spectrum of
bacteriohodOpsin which indicated that the protein was responsible for
the generation of the photovoltage. The photopotential was attributed
to the proton pump activity which is associated with the
bacteriorhodopsin protein (for review see Stoeckenius, Lozier and
Bogomolni, 1978). In these studies, however, it appears that the
bacteriorhodopsin protein did not become incorporated into the thick

planar membrane (Drachev et 21., 1976a). Following the same approach,

19

Shieh and Packer (1976) obtained similar results using planar bilayer
lipid membranes stabilized with polysterene. This same approach was
also used by Herrmann and Rayfield (1976) and Higgins, Lopez and Tien
(1979) (see Appendix 2) using BLM and also by Blok and Van Dam (1978)
who substituted the planar membrane with a lipid-impregnated millipore
filter. Another approach which was first used by Dancshazy gt al.,
(1976) and then adopted by others (Bamberg, Apell, Dencher, Sperling,
Stieve and Lauger, 1979) consisted of adding the purple membrane
patches to one side of a positively charged BLM. After some time the
system developed photopotentials in response to light due to the
adsorption of the purple membrane patches to the planar membranes. As
in all other studies, the action Spectrum of the photoresponses
followed the absorption Spectrum of bacteriorhodopsin. However, as in
previous studies the fact that the steady-state photocurrent increases
when the lipid bilayer is supplemented with proton conductors such as
CCCP and gramicidin A (Bamberg gt al., 1979; Herrmann gt 21., 1978)
suggests that the protein does not become actually incorporated into
the BLM but only becomes associated with the membrane in a preferential
orientation. Up to the present, there is no report in the literature
of a genuine incorporation of bacteriorhodopsin into a planar bilayer
where bacteriohodopsin regains transmembrane orientation analogous to
that existent in the bacterial membrane.

Another membrane component which has been reconstituted in planar
membrane systems iS rhodopsin, the visual pigment involved in the
primary event in visual excitation. It is believed that light-induced
conformational changes in rhodopsin eventually lead to cellular

excitation by modulating the ionic conductance of the photoreceptor

20

cell membrane (Haggins, 1972). However, the coupling mechanism of the

rhodopsin photochemical changes to the cell electrical response is not

known (Hubbell and Bownds, 1979). By incorporating rhodopsin in planar
membranes insight into this problem may be obtained.

Different rhodopsin preparations including rhodopsin-containing
vesicles, detergent solubilized rhodopsin and purified rhodopsin have
been used for reconstitution studies in planar membranes.

In studies by Montal and his collaborators planar bilayers
containing rhodopsin were formed by apposition of two preformed
monolayers (Montal, Darszon and Trissl, 1977). The monolayers were
formed by Spreading at an air-water interface a mixture made up of an
organic solvent (hexane or diethyl ether), rhodopsin and phospholipids.
They found that the membrane conductance increased with illumination.
The increase was to be irreversible and to have a latency period from
one up to several hundred seconds. This increase in conductance was
attributed to the light induced formation of channels in the
rhodopsin-containing bilayers. The same group apposed a monolayer
containing rhodopsin one side of a thin 16 um Teflon film separating
two aqueous compartments (Trissl, Darszon and Montal, 1976). The
Teflon septum acts as an electrical insulator and couples the
compartments only capacitatively. Fast photoelectric signals were
recorded when the membranes were illuminated with light flashes of
short duration. The photosignals were attributed to capacitative
charge displacements in rhodOpSin as it changes conformations. These
photoresponses are comparable to the early receptor potential (ERP) of
photoreceptor cells which is also believed to originate from

capacitative charge displacements in rhodopsin (Cone, 1967; Hagins and

21

McGaughy, 1968).

In another model system consisting of squid outer segment vesicles
adsorbed onto a thick ('blOOO A) planar membrane formed with monoolein
and decane, Takagi and Kishimoto (1977) have also recorded photosignals
similar to the ERP. More recently Liu and Tien (1982) have also
observed photosignals Similar to the ERP in a model system consisting
of rod outer segment (ROS) vesicles adsorbed to one side of a BLM. The
kinetics of the rise and decay of the photosignals followed the
kinetics of the transitions from lumirhodopsin to metarhodopsin I (Meta
I) and from Meta I to metarhodopsin II (Meta II) respectively (Liu,
Tien, Lopez and McConnell, submitted for publication).

A Similar model system, BLM formed in the presence of sonicated
rod outer segments, had been previously used by Fesenko and Lyubarskiy
(1977). In their studies an increased in the conductance of the BLM in
response to a single light flash was observed. The conductance
increased to its maximum value within 60 nS at 17°C and decayed to a
somewhat higher steady-state conductance than the original dark level
in about 1 S. No definite conclusion could be made as to the mechanism
of the photoeffect Since it was not known if the rhodopsin had been
incorporated into the BLM or if ROS vesicles had just sorbed to one
side of the membrane.

One other report of reconstitution of rhodopsin in planar
membranes involves the sonication of ROS with phospholipids followed by
lyophylization and resuspension in decane (Antanavage, Chien, Ching,
Dunlap and Mueller, 1977). The BLM formed with this suspension exhibit
a transient light-induced conductance increase, with a rise time of 5

ns and selective for H+. The response was Optimum in the presence of

22

2% ethanol. This last condition raises the possibility that the
results might have been due to released retinal since rhodopsin can be
denatured by the ethanol. BLM containing retinal are known to exhibit
light-induced H+ specific conductance changes (Kobamoto gt 11.,

1971).

Up to the present, however, there is no report of a planar bilayer
model system containing rhodopsin Similar to the way it is present in
the natural membrane. The planar membranes containing rhodopsin
actually incorporated into them which have been prepared are all
symmetric.

Cytochrome c oxidase, the enzyme which catalyses the reduction of
molecular oxygen to water during respiration, has also been
incorporated in planar model membranes (for a detailed review see
Montal gt 31., 1981). In studies by Drachev §t_gl., (1974) vesicles
containing cytochrome oxidase were added to one side of a thick planar
membrane. The vesicles adsorbed to the thick film and upon addition of
cytochrome c and ascorbate a cyanide-sensitive transmembrane potential
was generated. The results were said to be in agreement with Mitchel's
idea of transmembrane electron flow in the cytochrome oxidase segment
of the respiratory chain (Drachev gt_gl., 1974). In studies by Montal
(1974) planar bilayer membranes were formed by apposing two lipid
monolayers containing cytochrome oxidase. The membranes developed a
potential when oxidized cytochrome c was added to one compartment which
was reversed or prevented by cyanide. The origin of the potential
difference was attributed to a transmembrane electron transfer reaction
(Montal, 1976). Similar results were obtained by Chien and Mueller
(1976), who extracted cytochrome oxidase with phospholipids into octane

23

or decane and formed BLMS with the mixture. Tredgold and Elgamal
(1978) added cytochrome oxidase and lipids to hexane and formed bilayer
membranes by two techniques, with the usual brushing technique and by
putting together two opposing monolayers containing the enzyme,
subsequently on top of a mylar sheet. In both cases addition of
cytochrome c and ascorbate generated membrane potentials which were
sensitive to cyanide. The question of the nature of the
photopotentials was not directly addressed by the authors (Tredgold gt
31., 1978). Recently there has been a suggestion that cytochrome
oxidase may work as a redox-linked proton pump (Wikstrom and Krab,
1979). This view is contrary to the electron-translocating model of
the enzyme as proposed by Mitchel (1968). The use of planar bilayer
membranes may help resolve this question.

The first reports on reconstitution of photosynthetic reaction
centers from plants and bacteria in planar membranes were published by
Skulachev and collaborators in 1975 (Drachev, Kondrashin, Samuilov,
1975; Barsky, gt_al., 1976; Drachev gt al., 1976). In their studies
isolated reaction centers from both plants and bacteria were
incorporated into liposomes which were then added to one side of a
thick (not bilayer) planar phospholipid membrane in the presence of
divalent cations. The proteoliposomes containing bacteriochlorophyll

RC complexes from Rhodospirillum rubrum associated with the planar

 

membranes and photopotentials were observed in response to
illumination. The polarity of the light-induced potentials was
consistent with the proteoliposomes charging negative inside. The
photoelectric effect was shown to increase on addition of

tetramethyl-p-phenylenediamine (TMPD), coenzyme 05 (C006), and

24

vitamin K3 and to decrease on addition of fericyanide,
o-phenanthroline, and tricholorocarbonylcyanide phenylhydrazon (CCCP).
All the agents were added to the same Side as the proteoliposomes.
This group also did studies with plant chlorophyl complexes of P51,
however, no direct measurement of light-induced electric generation
were obtained (Barsky gt 31., 1976). The generation of an electric
field across the proteoliposomal membrane was demonstrated in this case
by means of the anion probe phenyldicarbaundecaborane (PCB'). The
effect required addition of phenazinemethosulfate (PMS) and its
polarity was consistent with the liposomes charging positive inside.
Photopotentials were also obtained in studies where R. rubrum
chromatophores were added to one side of a thick planar membrane. The
photoresponse required TMPD (or PMS) and could be enhanced by
naphtaquinone and ascorbate. In all of these studies the action
spectrum of the photoresponse followed the absorption spectrum of the
reaction centers. It was suggested that perhaps proteoliposomes with
RCS from plants could not associate with the planar membrane or that
electron transport in associated proteoliposomes could be inactivated
by decane which was present in the planar membrane.

RC from Rhodopseudomonas Sphaeroides extracted in hexane were

 

mixed with phospholipids and used to formed monolayers which were then
assembled into bilayers. The bilayers generated photovoltages and
photocurrents when supplemented with secondary donors, such as
cytochrome c, and acceptors as ubiquinone-o, on opposite sides of the
membrane. The photocurrent was biphasic, followed the absorption

spectrum of the RC5 and could be abolished by o-phenantroline.

25

Similar results were obtained by Packham gt 21., (1980) using
bacterial RCS extracted into octane. In this case the RC-octane
mixture was used to form BLMS by the brushing technique. Two kinds of
membranes were obtained "black" (bilayer) and thick (> 500 A). No
consistent difference was observed in the photoresponses exhibited by
the two kinds of planar membranes (Packham gt al., 1980). The RCs in
the phospholipid octane solution were shown to retain photochemical
activity. In both cases the generated photoelectric signals were
attributed to the transfer of electrons from a secondary donor to a
secondary acceptor located on opposite sides of the membrane (Packham

£31., 1980; Schonfeld _e_t__a_]_.,1979).

THE PHOTOSYNTHETIC APPARATUS OF PLANTS

(a) The chloroplasts:

 

In higher plants the process of photosynthesis takes place within
specialized cell organelles called chloroplasts. These chloroplasts
exhibit a characteristic lens shape with a long axis of 5 to 10 um when
looked under the microscope. However, when sectioned in a plane
perpendicular to the lens shape, the chloroplasts appear discoid. They
are surrounded by two non-pigmented membranes (inner and outer) which
enclose the highly organized internal lamellar structure and the
amorphous stroma. The lamellar inner structures are often referred to
as thylakoids. Usually a major part of the thylakoids is oriented in
an appressed form to constitute granum stacks or grana lamellae.

Single thylakoids usually connecting granum stacks are called stroma
lamellae. It has been preposed (Heslop-Harrison, 1966) that the

complex internal system of lamellae divides the volume inside the

26

chloroplasts into two compartments, each forming a separate continium.
It is suggested that the two compartments are formed by the folds and
connections of a single continuous sheet of membrane (lamellae). One
compartment is the stroma and the other represents the inner thylakoid
volume. This view is supported by the realization that the intergranal
lamellae were perforated, forming an interconnecting fretwork system
(Heslop-Harrison, 1963; Meier and Thomson, 1962) with layers connected
to lamellae at several levels within the same granum (Weier, Stocking,
Thomson and Drever, 1963). Some good reviews and monograms on the
structure and function of chloroplasts have been published and the
reader is referred to those for more detailed information. These
include descriptions of the structure of mature chloroplasts (Thomson,
1974), relationship between structure and photosynthetic function
(Arntzen and Briantais, 1975; Park and Sane, 1971) and aspects of
chloroplast development (Kirk, 1971). Also a comprehensive review has
been published by Gunning and Steer (1975). The more Specialized
function of chloroplasts in providing and organizing the machinery for
converting light energy into useful biological energy is associated
with the internal lamellar systems of the chloroplasts (Park and Sane,
1971). On the other hand, the dark biochemical reactions of C02 are
dependent on the soluble proteins of the stroma.

(b) The thylakoid membrane:

 

The internal lamellar system of the chloroplasts is the one that
contains the pigments, proteins and other molecules responsible for the
photochemical conversion of light energy into chemical energy. When
examined with the electron microscope at a high level of resolution the

appearance of the thylakoid membranes depends to a large extent on the

27

fixation procedure and the electron-opaque materials used in sample
preparation. In general, the various images obtained are similar to
those recorded for most membranes, ranging from the tri-partite (unit)
membrane structure consisting of two electron-dense regions about 2 nm
thick with a central translucent region of comparable width to a
membrane composed of globular subunits (Hohl and Hepton, 1965;
Murakami, 1964; Weier, Engelbrecht, Harrison and Risley, 1965). X-ray
structure studies on the thylakoid membrane are consistent with the
tri-partite model (Kreutz, 1969). Using electron microscopy and the
techniques of freeze-etch and freeze-fracture, it has been possible to
obtain information on the surface contours of thylakoids, and the
contours of fractured planes between the inner and outer membrane
surfaces. Four regions of the thylakoid membrane can thus be exposed:
inner and outer surfaces, and two complementary faces of an interface
generated by fracture. With these techniques particles of about 5 nm
in diameter can be observed attached to the outer surfaces (A' faces)
of the lamellar system (unstacked region). These particles however,
can be removed by washing the thylakoids with EDTA and buffers of low
ionic strength prior to freeze-etching (Arntzen, Dilley and Crane,
1969; Park and Pheichofer, 1968). They have been associated with the
enzyme ribulose diphosphate carboxylase (RuDP carboxylase) which is
responsible for C02 fixation and the CFl, which is the main part of
the enzyme responsible for synthesis of ATP from ADP and phosphoric
acid in photosynthesis (for a review see McCarty, 1979). The particles
are scarce or absent in the stacked regions outer surfaces (A). The
membranes are assymetric and when fractured exhibit different types and

number of particles on each face. In the unstacked region one finds

28

many bumps on the 8' face (fracture face outer lamellae), roughly of
two sizes, about 8 and 11 nm in diameter. The fracture face inner
thylakoid Side (0' face) has as expected the complementary appearance
of many pits and a few 11 nm bumps. In stacked regions the B ace
exhibits large bumps of about 16 nm in diameter, plus some 11 nm
bumps. The C face has large pits (complementary to the large bumps of
the 8 face), and also some small 8 nm bumps. The inner lamellar
surface of both stacked and unstacked thylakoids (D' and D faces)
exhibit medium and large bulges which may be reflections of bodies seen
on the fractured faces as well as other components of the
photosynthetic apparatus. In general, the results indicate that the
large bodies (16 nm) are found in the stacked regions and the small (8
and 11 nm) bodies exist in both stacked (granal) and unstacked
(stromal) lamellae.

Since the discovery that photosynthesis in plants involves the
cooperation of two photosystems, many investigators began trying to
isolate Specific components or fractions identifiable with one or other
photosystem. Using disruptive techniques like sonication, detergents,
french press and others plus differential centrifugation it has been
possible to obtain two fractions, a light buoyant fraction (Fraction 1)
enriched in photosystem 1 (PSI) activity and a dense heavy fraction
(Fraction II) having both activities but enriched for photosystem 2
(P52) activity (Boadman and Anderson, 1964; Michel and Michel-Wowertz,
1969; Ke and Vernon, 1967; Anderson, 1975; Becker, Shefner and Gross,
1965). When the fractions were examined with the electron microscope
it was observed the Fraction I consisted of small vesicles with
surfaces resembling those of stroma lamellae, while Fraction II looked

more like pieces of membrane in the stacked region, the grana lamellae.

29

In general, Fraction I contains 40 chlorOphyll a (long wave form)
molecules per P700 (photosystem I pigment) and no chlorophyll b.
Fraction II contains some P700, long and shorter wave forms of Chl a
and a ratio of Chl a to Chl b of less than 3. By subjecting the
fractions to further analytical treatments Thornber and his
collaborators (1976) discovered the light harvesting complex LHa/b, a
major antenna pigment-protein complex which contains equimolar amounts
of chlorophylls a and b and as mentioned before, has no photochemical
activity. In summary, the studies on Fractions I and II suggest that
thylakoid membranes contain three principal components: a major
antenna complex (LHa/b) and a component for each photosystem,
consisting of reaction center and a "subantenna" of Chl a. Several
polypeptides have been associated with the PSI activity of chloroplasts
having stacked regions (grana), P52 activity is confined mainly to
fractions derived from these regions; however, as yet, no morphological
entity in the membrane has been identified decisively with the reaction
centers P51 and P52.

(c) Charge separation and transport of electrons and protons in

plant photosynthesis:

 

The most widely accepted model of photosynthesis in higher plants
and algae is the so-called Z-Scheme of Hill and Bendall (1960). It
proposed two light-reactions in series activating the transport of
electrons from H20 to nicotinamide adenine dinucleotide phosphate
(NADP) via changes in the redox potentials of components of an
interconnecting electron transport chain. In all, except the two
light-driven steps, the flow of electrons is exergonic, in the

direction of redox equilibrium and constrained by the relative position

30

of the electron transport intermediates in the membrane. Several
detailed reviews on this model have been published by Vernon and Avron
(1965). Hind and Olson (1968), Bishop (1971) and Bishop (1974).

In essence, the process is as follows. Light energy is absorbed
by the light harvesting system of P52 and trapped at a reaction centre
thought to contain a specialized form of chlorophyll, P680 (Doring,
Renger, Vater and Witt, 1969). Through charge separation the excited
P680 (P680*) donates an electron to a primary electron acceptor
Q(E'O=O.O V) reducing it and producing a strong oxidant, Z,
(Eo'=+ 0.8 V) capable of oxidizing H20 to 02. The identities of
the primary electron acceptors and donors of P51 (0 and 2 respectively)
have not been clearly established yet. There is some evidence which
suggest that 0 may be a quinone (Van Gorkom, 1974; Bensasson and Land,
1973). There is however, some evidence (Van Best and Duysens, 1977)
that still another earlier acceptor (referred to as I in the litera-
ture) may be present between P680* and Q, and it has been suggested
that pheophytin (Pheo a) plays this role (Klimov, Klevanik, Shuvalov
and Krashnovsky, 1977). Very little is known about the entities that
carry electrons from H20 and act as donors to oxidized P680 but the
process is not a direct reaction between P680+ and H20 (Amesz and
Duysens, 1977). It is a controlled sequence in which the HT and 02
liberated from H20 are deposited inside the thylakoids and the 02
diffuses out. One thing that is known however, is that manganese
(MnZT) is essential for photosynthetic oxygen evolution (Cheniae
and Martin, 1972; Den Haan, Gorter de Vries and Duysens, 1976).

As in the case of P52, light absorbed in the light-harvesting

system of P51 is trapped at a reaction centre, in this case believed to

31

contain a photooxidisable form of chlorophyll, P700 (Kok, 1956; Ke,
1973). Charge separation at this reaction center leads to the oxida-
tion of P700 (Eo'=+ 0.4 V) and the reduction of a primary electron
acceptor referred to as A1 (Eo' - 0.6 V) in the literature.
Another acceptor usually referred to as X in the literature was
believed to be the primary acceptor, however, some recent evidence
suggests there is still one earlier acceptor A1 (Shuvalov, Dolan and
Ke, 1979). The reduced form of Q or I and the oxidized form of P700
are re-oxidized and re-reduced respectively by electron transport via
the electron transport chain connecting the two photosystems. The net
results of the simultaneous operation of the two photosystems is then
the transfer of electrons from H20 to A1. Reduced A1 can, in
turn, be re-oxidized by the transfer of electrons to X (referred to
also as A2) then to A and B (collectively referred to as P430) and
from here to ferredoxin, ferredoxin-NADP reductase and finally NADP
(noncyclic mode). The electrons can also be transferred from FD to
cytochrome b6 (Cyt b6), also known as cytochrome 563 (Cyt 563), to
the plastoquinone (PQ) pool and back to oxidized P700 (P700+) in what
is known as cyclic electron transfer. The ferredoxins are
Fe-S-proteins that appear in many pathways of electron transport,
especially in anaerobic (not necessarily photosynthetic) bacteria.
Electron transport in the pathway interconnecting the two
photosystems is thought to take place in the following way. Electrons
from the primary electron acceptor of P52, 0 or I is transferred to the
plastoquinone (PQ) pool and then to cytochrome f (Cyt f).
Plastoquinone is the name given to a mixture of closely related

electron transport intermediates whose principal component is

 

 

" 0.8 ’ A1
p I (x) 32 H
E , l (7) AID
7-" _ _ ‘* Reduc- y.
‘3 0'4 NADP“ .. tase FD
a. >
E » Qa ./
3. L \ cyg
3.; 0|) /563
'. 0 ‘ \Po
§ » Pool
5 I
.8 h \m
g b M‘ h'
" . C t f»
E I 0'4 y "’8 P700
‘1 .
fl-
‘ I-
M

b

1' 0,8 )- “20% z
1

Figure 2.1 - Z-scheme of plant photosynthesis.

*P680

 

Current view of the so-called Z-scheme for green plant
photosynthesis showing the paths of electron transfer in
the R05. The ordinate Shows the approximate redox
midpoint potential of each component. The question marks
indicate the tentative nature of the assignments.

33

plastoquinone A. The concentration of this type of compound in chloro-
plasts is higher than that of any other electron transport intermediate
(Amesz and Duysens, 1977). The concentration is normally about 5-10%
of the total chlorophyll or approximately 25-50 molecules per P51 or
P52 unit. There is some evidence that another electron transport
intermediate, M, may be located between the PO pool and cyt f (Levine,
1969). In studies with duckweed chloroplasts, Malkin and Aparicio
(1975) have observed a light dependent ESR Signal that could be due to
the hypothetical M. The Spectrum of the signal suggests that it is an
Fe-S center. From cyto f electrons are then transferred to plasto-
cyanin (PCy), a copper-protein compound, and then to oxidized P700.

Linked to this flow of electrons through the electron transport
chain there is also a transport of protons whose possible role in
photosynthesis will be discussed later (for review see Crofts and Dood,
1978). This transport takes place as protons are bound from outside
the thylakoid when PO is reduced, and released inside when PO is
reoxidized. This results in a net transfer of H+ from outside the
thylakoids to inside. The evolution of 02 is also accompanied by a
net release of H+ inside the thylakoids (Saphon and Crofts, 1977).

Most of the information given above on the involvement of
photochemistry and proton and electron transport in photosynthesis has
come mainly from information of light induced and chemically induced
changes in absorption and emission of light and ESR (see for example,
Babcok and Sauer, 1975; Evans, 1977; Malkin, 1977; Mathis, 1977;
Pulles, Van Gorkom and Willensen, 1976; de Grooth, van Grondelle,

Romijn and Pulles, 1978). These changes usually signal the oxidation

34

or reduction of molecules. The studies also involve manipulations of
the material under investigation, as for example: (1) the use of
intense Short flashes of actinic light, (2) selection of the
illumination wavelength so as to drive either P52 preferentially, (3)
fractionation of chloroplasts into particles enriched in P51 or P52
activity, (4) adjustment of the ambient redox potential, pH and
temperature, (5) deletion of selected components by the use of specific
inhibitors, genetic mutation and extraction, (6) replacement of

deleted components or substitution with analogous, or addition of
artificial electron donors and acceptors.

(d) Electric potential across the thylakoid membrane and ATP

synthesis:

The most important problem in ATP synthesis involves conversion of
the redox energy of electron transport into the anhydride bond of ATP.
Three of the hypothesis which have been proposed involve a high energy
intermediate to connect electron transport with ATP formation and are
therefore known as coupling hypothesis. The difference between them
lies in the nature of the high energy intermediate.

The chemical coupling hypothesis (Chance and Williams, 1956)
proposes that a molecule or molecular complex plays the role of the
high energy intermediate. The complex has the pr0perty that the
affinity for H20 is changed by a change in the redox state of the
electron carrying part and the redox potential of the latter depends
on the state of hydration. This hypothesis was successful in
accounting for the formation of ATP linked to the fermentation of

sugars and other organic compounds through soluble components.

35

In the conformational coupling hypothesis (Boyer, 1974) the energy
intermediate are stresses induced by the transport of electrons which
causes changes in the conformation of macromolecules. Due to these
changes, in turn, the equilibrium between hydrolysis and dehydration is
shifted to favor ATP formation.

This last hypothesis to be mentioned is the chemiosmotic
hypothesis of Mitchel (1961), which proposes that the high energy
coupling intermediate is a thermodynamic state equivalent to a proton
concentration cell. A protonmotive force (pmf) established across the
thylakoid membrane (in the case of plant photosynthesis) by the
light-driven translocation of H+ from outside to inside drives the
movement of protons through an ATP-forming enzyme (ATPase) and ATP is
synthesized.

This hypothesis to explain photosynthetic phosphorylation is the
only one which offers an explanation to the observation that both
oxidative aS well as photosynthetic phosphorylation required an intact
vesicular membraneous structure separating two aqueous regions. It
also offers a very plausible explanation for the degree of chemical
diversity of those substances which interfere with phosphorylation and
are known as uncouplers. In terms of the chemiosmotic hypothesis
uncouplers act by allowing the passive movement of H+ and other ions
dissipating the pH differential and/or the membrane potential.

Evidence has been obtained that there is a relationship between
ATP synthesis and H+ and electrical gradients. First comes the
classical and dramatic experiment of Jagendorf and Uribe (1966) which
Showed that ATP could be synthesized, without the aid of light, with an

artificially imposed pH gradient across the thylakoid membrane. In

36

this experiment chloroplasts were added to a medium containing succinic
acid (permeates thylakoid membrane) at pH 4 in order to lower the pH
inside the thylakoids to 4. The external pH was then suddenly raised
‘ to 8 resulting in ATP formation. Then comes the experiment by Miles
and Jagendorf (1969) who Showed that ATP synthesis and light—induced
proton uptake by chloroplasts have the same kinetics. More recently
Graber, Schlodder and Witt (1977) showed that the membrane potential,
the other component of the pmf, can also yield significant ATP
formation when imposed artificially. In this experiment chloroplasts
were suspended between electrodes 1 mm apart with a potential of 200 V
applied across the electrodes. The artificial electric gradient
which was estimated to produce a gradient of about .3 V across the
thylakoid membrane resulted in ATP formation.

In another type of experiment, not involving thylakoid membranes
however, Racker and Stoeckenius (1974) prepared lipid vesicles

containing purple membrane from H. halobium and bovine heart

 

mitochondrial ATPase. When the vesicles were illuminated (which
according to Racker and Hindle, 1974, and Kayushin and Skulachev, 1974,
Should induce both a membrane potential and pH gradient) ATP was
formed. These experiments were repeated by Ryrie and Blackmoore (1976)
with yeast ATPase, by Winget, Konner and Racker (1977) with ATPase from
spinach chloroplasts and by Yoshida, Sone, Hirata and Kagawa (1975)
with cristalline ATPase from the thermophillic bacterium PS 3.
Electrogenic potential generation is expected to occur in
association with a charge separation across a membrane. It has been
suggested (Junge and Witt, 1968) that such a charge separation takes

place across the thylakoid membrane as a result of the excitation by

37

light of the photosynthetic reaction centers present in the membrane.
Some evidence that such a potential difference developes across the
thylakoid membrane in response to illumination has been obtained.
Several methods have been applied to demonstrate the primary
electrogeneis at the thylakoid membrane upon light excitation. These
include: 1) the fast light-induced absorbance change at 515 nm, first
discovered by Duysens (1954), later observed also in photosynthetic
bacteria (Amesz and Vrendenberg, 1966). The change in absorbance has
been attributed to a shift in the absorption bands of the native
pigments in the membrane in response to an electric field
(electrochromic shift) by Junge and Witt (I968). The rise time of the
change was found to be less than 20 ns (Wolff, Buchwald, Ruffel, Witt
and Witt, 1969). Spectral shifts of carotenoid, similar to the ones
observed in photosynthetic bacteria, were observed in valinomycin
treated chromatophore preparations in response to a KCl-induced
membranes potential (Jackson and Crofts, 1969). 2) Charge
polarization as measured by external electrodes in a layer of stripped
chloroplast which was illuminated by a beam of light perpendicular on
the layer (Witt and Zickler, 1973). 3) Several attempts have been
made to associate the delayed fluorescence first observed by Strehler
and Arnold in 1951 with the generation of a membrane potential
(Fleischmann, 1971; Crofts, Wraight and Fleischmann, 1971; Barber and
Kraan, 1970). This delayed light emission is believed to be the
reversal of the initial photochemical and thermochemical events associ-
ated with P52 (Strehler gt_gl. 1951) and arise as a result of charge
recombination within that reaction center (Arthus and Strehler, 1957).

It has been suggested that the magnitude and intensity of ms delayed

38

light emission was an expotential function of the membrane potential
(Barber and Kraan, 1970). Bell, Haug and Good, (1978) however, have
questioned this interpretation based on the results of their
experiments where they observed stimulation and not inhibition of
delayed fluorescence by uncouplers when the latter caused an increase
in the rate of electron transport. Their data suggested that the
decrease in intensity of millisecond-delayed fluorescence was not due
to the elimination of a membrane potential but “more likely a
reflection of the rate of disappearance of some other electron
transport-generated condition, a condition which is
uncouplers-insensitive". 4) A more direct method has been employed by
Bulychev, Andrianov, Kurella and Litvin (1972) who used microcapillary
electrodes to measure the light induced potential difference in
chloroplasts. This approach although the most desirable, has the
dissadvantage that the position of the electrode inside the chloroplast
is not known, and if inside a granumm stack, there is the possibility
of inducing ionic leakage. The kinetics of the 515 nm absorption
change in illumination periods of 1 s or more, which have been
suggested to indicate a proportional change in the membrane potential
(Junge gt 11., 1968) are different from those measured with
micro-capillary glass electrode. 5) One other method which has been
applied successfully in the photosynthetic bacteria involves the
reconsitution of the photoactive membrane or the reaction centers (RCs)
from the membranes into artificial planar model membranes (Skulachev,
1979; Schonfeld §£_§l. 1979; Packham gt 11., 1980). This method has
the advantage that it allows for direct measurement of the potential

and some control over the conditions under which the potential will be

39

generated. The main disadvantage of this approach lies in the
limitations imposed on the reconstitution by the characterisitics and
properties of the artificial membrane system. It is expected that for
successful reconstitution the functional activity of the system Should
be maintained and no artifactual behavior should be induced. This
approach has not yet been applied successfully to the RC5 of plants.
In the experiments with bacterial RCS different reconstitution methods
were used. Skulachev's group used the fusion approach, where vesicles
containing bacterial RCs were added to one Side of a thick (3_IOOO A)
planar membrane so that they fuse with it (for a review see Skulachev,
1979). Schoenfeld gt_al., (1979) obtained a reaction center-lipid
complex in hexane and formed planar bilayers by apposing two monolayers
assembled from lipids and reaction center-lipid complexes in hexane.
Packham EE.El-: (1980) formed planar membranes (thickness > 500 A) by
brushing into a teflon septum with a small aperture, a membrane forming
solution containing bacterial RCS. The RCs were incorporated into the
octane containing forming solution using phospholipids to carry the RC5
into solution.

In all cases, photovoltages and photocurrents were obtained when
the systems were supplemented with secondary donors and acceptors. The

photoresponses all follow the absorption Spectrum of the RC5.

CHAPTER 3

EXPERIMENTAL METHODS AND MATERIALS

MATERIALS

(a) Chemicals:

The chemicals used in these studies were obtained from chemical
companies and used without further purification, with the exception of
cholesterol which was recrystallized three times in ethanol. The
chemicals include: n-butanol, calcium chloride (CaClz), carbonyl
cyanide m-chlorOphenylhydrazone (CCCP), chloroform, cholesterol,
a-chymotrypsin, coenzyme 05 (C005), n-decane, n-dodecane,
ethylenediaminetetraacetic acid (EDTA), ferric chloride (FeCl3,
hydroxylamine (NHZOH), magnessium sulfate (MgSO4-7H20),
N-bromosuccinimide (NBS), n-octane, o-phenanthroline, phenazine
methosulfate (PMS), egg phosphatidylcholine (PC), bacterial
phosphatidylethanolamine (PE), bovine phosphatidylserine (PS),
potassium chloride (KCl), potassium carbonate (K2C03), sodium
acetate (C2H3Na02), sodium ascorbate, sodium borohydride
(NaBH4), sodium chloride (NaCl), sucrose, N-tris
(hydroxymethyl)-methylglycine (Tricine), trisodium citrate, vitamin
K3 (Vit K3, (Difco) Yeast Extract.

(b) Composition of solution used for forming BLM:

 

In most cases (except where specified) the BLM-forming solution

40

41

consisted of a mixture of phospholipids and cholesterol dissolved in an
n-alkane. The composition is as follows: 1.2% egg phosphatidylcholine,
2.2% bacterial phosphatidylethanolamine, 0.7% bovine phosphatidylserine
and 0.8% cholesterol in n-octane or n-decane. All percentages are
weight by volume. In the experiments on incorporation of thylakoid
membrane fragments in BLM the cholesterol was omitted. In some
experiments on photoelectrospectrometry of BLM (see Appendix A) a
BLM-forming solution made-up of an extract from spinach leaves
dissolved in a 1:1 mixture of n-butanol and n-dodecane was used.

(c) Purple membrane (PM):

The PM used in these studies was extracted (see Methods section d)

from Halobacterium halobium cells strain R1 grown in the laboratory.

 

A sample of the cells was originally obtained as a generous gift from
Professor Cassim of Ohio State University.

(d) Thylakoid membrane fragments (TMFs):

The TMFS used were obtained from spinach chloroplasts following
published procedures (see Methods section 9). The spinach was obtained
from local markets.

(e) Experimental set-up for electrical measurements and formation

w:

A diagram showing the experimental arrangement used in the BLM
experiments is shown in Figure 3.1 It consists of:

(1) Light source- 1000 W Hannovia Xenon gas lamp with Schoeffel

Instruments power supply.
(2) Monochromator A Baush and Lamb 250 mm light path
monochromator equipped with a 600 grooves/mm difraction

grating was used together with the light source to illuminate

(9)
(10)

(11)

(12)

(13)

42

the membrane with light of different wavelengths.

Motor - A dc variable speed motor used to drive the grating.
Variable voltage source - A box containing two 1.5 volts Size
0 batteries and a variable resistance were used as a variable
source of power for the motor.

Shutter - Functions to control the duration of illumination.
Lens - Quartz lens to focus the light on the BLM.

BLM outer chamber - Consists of a plexiglass block with two
holds bored through it (one of them playing the role of outer
chamber) with two windows, one for illumination (quartz) and
the other for observation (glass).

Teflon septum for BLM formation - A 10 ml Teflon beaker with
a 1.5 mm in diameter hole punched through it.

Stirrers - Two magnetic stirrers one in each chamber.

Calomel electrodes - Two colomel electrodes were used to
provide electrical contact with the membranes. The inner
chamber electrode was anode (or high impedance electrode) and
the outer chamber electrode was cathode (or reference
electrode).

Electrometer - Keithley 6IOBR electrometer used to measure
the potential difference across the membrane.

Picoammeter - Keithley 417 high speed picoammeter used to
measure the current flowing through the membrane.

External variable voltage source - A box containing a simple
circuit consisting mainly of a 1.5 V size 0 battery and a 10
turn helipot used as external variable source of voltage and

current for the BLM.

43

(14) Chart recorder - Baush and Lomb VOM6 recorder used to record
the time course of the photovoltages and photocurrents.

(15) X-Y recorder - MFE Plotamatic 715 M X-Y recorder was used to
record the action spectrum of the photoelectric effects.

(16) Capacitance meter - ICE/Electronic Model 1-6 low level
capacitance meter used to measure the BLM capacitance.

(17) Flash unit - General Radio 1538A Strobotac flash light used
to measure the photoresponse to short duration illumination.

(18) Oscilloscope - Tektronic R5031 Dual Beam Storage oscilloscope
used to record the photoresponse to short duration
illumination.

(19) Special circuit for measuring the photoresponse to Short
duration illumination.

(20) Special circuit to monitor changes in BLM conductance (see

Figure 3.2).

METHODS
(a) Formation of BLM:

 

BLMS were formed by the usual technique of injecting a small
amount of "forming" solution with the aid of a Hamilton mycrosyringe,
over a small (1.5 mm in diameter) hole located on a Teflon septum (see
Tien, 1974). In this case the Teflon septum consited of a 10 ml beaker
located in a plexiglass chamber with two mutually perpendicular
windows, one made of quartz, for illuminating the membrane, and the
other made of glass for observing the membrane. In all experiments
addition of chemicals or membrane preparations was done after the

membranes had reached the bilayer ("black") stage.

44

.mmwusum zgm mg» C. vow: “cosmmcmccm Poucwe_cwaxm any we Ewcmmwu xoo—m . H.m wczmru

 

45

 

 

V

3.3 9.363 as... :3:

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cottage“ :95
QED; ox
BEES?
IOUE flm on? . ‘\
no .0...
A: .m: 3: 5:25
3:23.33 2.5 0).!) ..anoiuocoz
of mctohcoE . 0:5 2.3
cozoh
50* :35. .2935 <
— .o o
F i. . g
I AI u o
I mowoctofi 02.0 x
—
._- Lag

 

 

 

 

 

 

, 3.30m
xom 20.293. 5.833. >-x .5sz .202

amino: .3 cozofixo o. o .
omconmw. 9t mESmowE x .
50.055003... T /,_\(/\ 50—92308 top—U
.532

.0. :32? tutonm DB rm
coonzuoE
oucuzuunou 6

__ : A. >
I i .. .\

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

2330530 I

 

 

 

 

 

 

 

 

 

E=L5V

 

 

 

 

 

 

 

to Recorder
6

 

 

Figure 3.2 - Schematic of the circuit used to monitor the BLM
conductance.

The value of R R2 and R3 were chosen so that a
change in RT oi up 2to two3 orders of magnitude would only
cause a sma Tl change in the potential difference across
the BLM. (R1= 820 a, R2= O - 100 3, R3 = 6 a.

47

(b) Culture conditions for H. halobium:

 

The cells were grown under intense illumination (J «.5 x 103
erg/cmz-s) from a bank of fluorescent tubes at 37°C in two-liter
flasks located inside an incubator. The growth medium per liter of
culture consisted of: NaCl, 250-0 9; MgSO4-7H20, 20-0 9;

KCl, 2-0 9; CaClz-ZHZO, 0.2 g; trisodium citrate-ZHZO,

3.0 g; Difco Yeast Extract (0127-03), 5.0 9. Each culture was
inoculated with 200 ml of cells from a previous culture. The cells
were aerated by bubbling air through the culture at a rate of 15 cubic
feet of air per hour for 5 days or until the end of log phase. Then
the cells were aerated at a rate of 8 cubic feet per hour for 3 days.
After that the cells were harvested.

(c) Harvesting:

 

For harvesting the cells were spun down by centrifugation at 8000
rpm for 15 minutes. The pellet collected this way was then washed
twice in basal salts (first 4 items listed above for growth medium) and
resuspended in basal salts.

(d) Isolation of purple membrane:

 

The purple membrane (PM) was isolated by dialyzing H. halobium

 

cells for 24 hours against 12 liters of 0.1 M NaCl (4 liter container
change 3 times). The fraction from the dialysis was then centrifuged
for 1 hour at 31,000 g. The supernatant was then decanted and 5 ml of
0.1 M NaCl were added to the pellets which were then gently shaken to
resuspend the PM while keeping the muddy-colored cell-debris as a
pellet. The resuspended PM was decanted and homogenize 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

48

orange to faint purple. The final pellet was resuspended in 10 ml of
double distilled water (dd H20) and then subjected to a high speed
density gradient centrifugation. A continuous 30%-50% sucrose
gradient, with a 1 ml 60% sucrose cushion, was prepared with 5 ml of
each 30% and 50% sucrose added to a density gradient maker. Two ml of
the pH sample were added to the gradient in each of Six cellulose
nitrate tubes. The gradients which were centrifuged to equilibrium for
18 hours at 35,000 rpm in a Beckman LB ultracentrifuge using an SWTi 41
rotor, yielded a dense purple band and a diffuse 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 centrifugation was
washed twice in dd H20 at 31,000 g for an hour and stored as A
pellets.

An absorption spectrum of the pM obtained with this procedure is
shown in Figure 3.3. The spectrun is similar to the one reported in
the literature (Oesterhelt and Stoeckenius, 1974), and exhibits a
ratio of 280 nm (protein) peak to 560 nm (chromophore) peak of 2.0.

(e) Chemical modification and/or bleaching of PM:

 

Three different types of chemical modification and bleaching of PM
were done: (1) chemical modification with N-bromosuccinimide (N85), (2)
bleaching with hydroxylamine (NHZOH), and (3) bleaching with sodium
borohydride (NaBH4).

(1) Chemical modification (with bleaching) of PM with NBS was
carried out following the method of Konishi and Packer (1977). Enough
PM for a bacteriorhodopsin (bR) concentration of 0.2 9/1 («.8 uM) was
suspended in 10 ml samples of 0.1 M KCl and 0.02 M Na-acetate at pH
4.7. To each 10 ml sample, a different amount of N85 was added. The

 

49

 

.xcoumconoF mgu cw czocm owcmpuwn soc» meanEme wracza mo Escuomam :owpqcomn< . m.m mesmwm

50

 

 

1.0 .

 

0.8 F

J I

 

 

o v

6 e
aoueqaosqv

0.2 .

700

500

400

300

(nm)

A

51

absorption Spectra of the samples containing the different molar ratios
of NBS to bacteriorhodopsin were measured and are shown in Figure 3.4.
The spectra agree with the ones obtained by Konishi £3.21- (1977).

(2) The method used for bleaching PM with NHZOH was a variation
of the ones reported in the literature (Oesterhelt gt 31., 1974; Becher
and Eberly, 1976; Bauer, Dencher and Heyn, 1976). The method is as
follows. A fresh solution of 0.2 M NHZOH at pH 7 was prepared and
divided into 20 ml samples. An amount of PM containing 4 mg of bR
protein was then added to each sample. Each sample was illuminated for
a different amount of time with light from a 1000 W Xenon lamp. The
light was filtered through 5 cm of a 1% CuSO4. After the period of
illumination for each sample had ended, the bleached PM was washed with
0.15 M KCl twice by centrifugation at 30,000 g for 1 hour and
resuspended in 0.15 M KCl. The absorption Spectrum for the bleached PM
was measured (see Figure 3.5) and found very Similar to the spectra
published by other authors (Oesterhelt gt 21., 1974).

(3) The method used for bleaching PM using NaBH4 is a variation
of the method used by Peters, Peters and Stoeckenius (1976). PM was
suspended in 10 ml samples of freshly prepared carbonate buffer, pH 10.
The concentration of bR in each 10 ml sample was 0.2 mg/ml ('o8 pH).
10 mg of NaBH4 were added, the samples were then placed in an ice
bath and illuminated with a bank of fluorescent tubes. Each vial was
removed at a different time. After all the samples were collected they
were washed several times by centrifugation and resuspension in 0.15 M
Na-phosphate buffer at pH 7. The process was repeated until the
residual NaBH4 was removed. The absorption spectrum of the bleached

PM was measured (see Figure 3.6) and found in agreement with the ones

52

 

1.0

Y

Absorbance

0.4,-

0.2

 

    

 

 

 

400 500 600 700
A (nm)

30M)

Figure 3.4 - Absorption spectra of purple membrane modified with NBS.

Each curve represents a different molar ratio of NBS to
bacteriorhodopsin (bR). The molar ratio is equal to: 4 in
curve (I), 10 in curve (2), 20 in curve (3), and 40 in

curve (4).

53

.mcso; mN Lo» umuocwE:_Fw z; .m m>c=u ”meson
N co» umumcwe:F—P :8 .N m>c=u mumumcw52~rp no: Azav mcmgnsme opaczg .H o>c=u

.29:
mo mucmmmca mg» cw zomzz saw: umgumm—n mcmcnsme wracsa eo ocuumam cowquomn< . m.m wcamwu

54

 

 

 

A L A

 

 

 

 

 

1.2

1.0 ..

A
Q GD
0
C G

0.4 ~

aoueqaosqv

0.2 I

700

500

400

300

("I")

A

55

.Ac m>g=uv awe mm van .Am m>c=ov c_s ma .AN m>g=ov are m .AH
m>c=ov are C ”Loy u;m_— op ummoaxm Ea pcommcaoc e op H mm>gsu .oH :a pm usmwy
can axmmz sue: Atmoaumc :ngowozcowcmpumnv umzumm_p mm: meccaems opacsa as»

.29:
$0 mocmmmca ms» cw axmmz guwz vmzummrn mcmcnsoe m—gcza mo wepuwam cowuacomn< . m.m mezmwu

 

 

 

 

 

 

 

aoueqaosqv

700

500

400

300

(lllll)

X

57

previously published (Peters gt 31., 1976).

(f) Isolation of chloroplasts from spinach leaves:

Between 80 and 100 g of whole green leaves were washed with
distilled water, had the stalks and ribs removed and were allowed to
dry for 30 min. The leaves were then chopped in a blender in 150 ml of
0.4 M sucrose plus 0.05 M Tricine-OH (pH 7.5) at low speed first and
then after all the leaves had been added, at high Speed for 30 S. The
homogenized mixture was then filtered through eight layers of cheese
cloth and the filtrate was collected. The green suspension was then
centrifuged at 200 g for 5 minutes and the supernatant was collected.
The supernatant was then centrifuged again for 12 minutes at 1000 g and
the pellet which contains the chloroplasts was collected. The
chloroplasts obtained with this procedure were then used for thylakoid
membrane extraction.

(9) Isolation of thylakoid membrane fragments (TMFS):

Two different procedures were used to isolate TMFS. The first
procedure was developed by Douce, Holtz and Benson (1973) and it
involves the rupture of chloroplasts by gentle osmotic shock. The
procedure is as follows. Intact purified chloroplasts in 0.4 M sucrose
3.0 ml, about 8 mg chlorophyll per ml, were treated for 2 minutes at
4 °C in 70 ml of swelling medium (10 mM Tricine-OH buffer, pH 7.5 and 4
mM MgC12. Samples of 5 ml were then layered on top of a
discontinuous sucrose gradients and centrifuged in a swinging bucket
rotor for 60 minutes at 20,000 rpm in a Beckman model L3-50
ultracentrifuge. The gradients were prepared by layering 3 ml of
sucrose solutions containing 5 mM Tricine-OH buffer at pH 7.5 and 2 mM

MgCl2 into six tubes in the following sequence: 1.5, 1.2, 0.93 and

58

0.6 M. The centrifugation resulted in a separation of the membranes
into three distinct bands which formed at the top of the inerfaces of
the sucrose layers. A dark green fraction (heavy fraction) which
sedimented a top the 1.5 M sucrose was collected from all 6 tubes using
a syringe. The collected material was combined, diluted four times
with 10 mM Tricine-OH buffer at pH 7.5 mixed well, and spun for 1 hour
at 27,000 rpm. The pellets were suspended in a medium containing 0.3 M
sucrose and 10 mM Tricine-OH buffer at pH 7.5. The chlorophylls
content was determined using the method of Arnon (1949), which is based
on the specific absorption coefficients of chlorophylls a and b in 80%
(V/V) acetone as determined by McKiney (1941).

The second method used is a variation of the one used by Becker,
Shefner and Gross (1965) and makes use of the more drastic technique of
sonication. The chloroplasts are resuspended in low ionic strength
medium, 15 mM KCl and 5 mM Tricine-OH at pH 7.5 and stirred for one
hour (low stirring) at 4 °C. The disrupted chlorplasts are then
sonicated in an ice-water bath with a Branson model 140-0 Sonifier Cell
Disruptor for a total of 9 minutes (3 times for 3 minutes with a 1
minute rest between sonication periods) at 60 W. The sonicated
chloroplasts were then centrifuged for 1 hour at 30,000 g. The
supernatant which consisted of thylakoid membrane fragments (TMFS) was
collected. As before, the chlorOphyll content was determined by the
method for Arnon (1949). It was found that the concentration of total
chlorophyll ranged beween 0.8 and 1.6 mg/ml per 100 g of leaves.
Although these numbers appear to vary considerably, it was also found
that the relative amount of Chl a and Chl b before and after

centrifugation were very similar from preparation to preparation.

59

(h) Preparation of liposomes or vesicles containing PM or TMF:

 

The method used to prepare liposome is essentially the method of
Blok_gt.al. (1978). 100 mg of PC in hexane or chloroform was poured
into a round bottom flask and the solvent was evaporated to dryness by
means of a Buchner flash evaporator. 10 ml of 0.15 M KCl and 0.01 M
Tricine-OH (pH 7.5) or 0.1 M Na-acetate (pH 5.0) were added, depending
on the particular experiment. The phospholipids were dispersed by
adding a few glass beads and shaking in a vortex mixer for 10 minutes.
The biological membrane to be incorporated was added and the mixture
was sonicated at 80 W for 15 minutes (60 sonication period of 15 S,
each followed by a 45 5 rest period) in an ice bath under a stream of
N2. The vesicles were then centrifuged for 30 minutes at 10,000 g to
remove heavy fragments and titanium particles.

In the case of PM-vesicles, the concentration of bR in the vesicle
suspension was 0.2 mg/ml (rv8 uM). For TMF-vesicles enough sonicated
broken chloroplast suspension was added to the phospholipid dispersion
for a total Chl concentration of 0.1 mM. In experiments with
TMF-vesicles containing C006 or Vit K3 in the vesicular membrane,
the concentration of the quinones in the vesicle suspension of 0.1 mM.
In experiments involving incorporation of TMF in BLM, the TMF vesicles
were prepared in a buffer solution supplemented with sucrose to a final
concentration of 0.2 M sucrose. Electron microscopy of the vesicles
revealed that they consisted of closed structures (see Figure 3.7).

(1) Fusion of vesicles with BLM:

 

Two different vesicle preparations were used for fusion studies:
PM-vesicles and TMF-vesicles. In both cases the same method was used

to achieve fusion. The BLM was formed with a solution which contained

60

Figure 3.7 - Electron micrographs of the vesicles used in these
studies. (34,560x) A) PM-vesicles. B) TMF-vesicles.

61

 

62

a negatively charged phospholipid, phoaphatidylethanolamine. A
divalent cation, usually Ca2+, was present in the bathing solution.
With both of these conditions met, addition of either vesicle
preparation resulted in fusion of the vesicles with the BLM.

(j) Incorporation of proteins into the BLM:

Two distinct approaches were used to incorporate proteins into
BLMS. One approach was to add the protein to the aqueous phase at one
side of the BLM and wait until it adsorbs onto the membrane. It was
used in the experiments with a-chymotrypsin (see Appendix A). The
other approach involves the incorporation by means of total fusion of
the vesicle with the BLM. This approach was used to incorporate TMF,
or more Specifically, PSI reaction center into the BLM. In order to
obtain total fusion between the vesicle and BLM, Ca2+ (to a final
concentration of 10 mM) had to be present on the side containing the
vesicles. Also, and perhaps more important, the vesicle interior had
to be hyperosmotic with respect to its surrounding aqueous environment.
This was accomplished by preparing the vesicles in a medium containing
0.2 M sucrose, 0.1 M KCl and 5 mM Tricine-OH at pH 7.5. The aqueous
phase on Opposite Sides of the membrane consisted of 0.1 M KCl and 10
mM Tricine-OH at pH 7.5 and 10 mM Ca2+ on the Side to which the
vesicles were added.

(k) Electrical measurements:

 

Electrical measurements were done as usual (see Tien, 1974). For
the action spectra measurements, the diffraction grating was moved
mannually in increments of 20-25 nm or continuously by means of a
motor. For the experiments where the conductivity of the BLM was

monitored, the circuit shown in Figure 3.2 was used. The circuit is by

63

no means a voltage clamp circuit; however, the changes in voltage

should be small as long as the membrane resistance stays high.

CHAPTER 4

PHOTOELECTROSPECTROMETRY OF BLM

INTRODUCTION

As mentioned previously, one approach used to study how the
components of biological systems work is to reassemble them into model
membranes, including BLMS and liposomes. Liposomes are very suitable
for studies involving spectroscopic techniques. This is mainly because
by being rather small they have a relatively large surface area to
volume ratio. Light spectroscopy studies (e.g. fluoresence,
absorption, etc.) on BLMS are rather difficult and usually require the
use of sophisticated equipment (see for example Steinemann, Stark and
Lauger, 1972). One way to overcome this difficulty is to do photo-
electric spectroscopy studies on the BLM. These involve the measure—
ment of the action spectrun of the photoelectric effect from a photo-
sensitive BLM and usually consists of a photovoltaic or phoconductivity
action spectrum. The technique itself is fairly Simple and due mostly
to the high electrical resistance of the BLM, rather sensitive.

In this chapter, the results of photoelectric spectroscopy studies
on BLMS modified with certain photoactive Species will be reported.
The purpose of the studies is two-fold: First, to explore the
possibility of extending such studies into the uv region, the region

where most proteins absorb radiation, and second, to examine the kind

64

65

of information which can be obtained from such studies.
The results are divided into two parts. The first part includes
the results of studies with BLMS containing chlorophyll pigments,

purple membrane (PM) from H. halobium and adsorbed chymotrypsin. Since

 

these results have already been published, the reader is referred to
Appendix A for details on the results and discussion. The second part
deals with a more detailed study of the PM-BLM, namely photoelectric
Spectroscopy (or photoelectrospectrometry) of chemically modified PM
reassembled into BLMS. Before going into the second part, however,
some recent observations relevant to the first part will be discussed.

In the experiments discussed in Appendix A, it is shown that
chymotrypsin molecules, when adsorbed onto one side of a BLM, can
generate photovoltage if irradiated with uv light. Chymotrypsin iS
normally a non-photoactive protein, however the initial charge
separation which leads to the observed photovoltages could be explained
in terms of photoionization of the aromatic amino acids by the uv
radiation. This is known to occur in biological compounds and iS
believed to be the major initial photoreaction in radiation damage
processes (Grossweiner, Brendzel and Blum, 1981).

After it was found that the action Spectrum followed the
absorption Spectrum of the aromatic amino acids, it was decided to see
if the aromatic amino acids adsorbed onto one side of a BLM would also
generate photovoltages. The experiments were done, but no
photovoltages were observed. More recently, however, Huebner, Arrieta
and Millar (1982) have reported the observation of transient
photovoltages in response to light flashes of 5 us duration from

aromatic amino acids adsorbed on one side of BLMS. They have also

66

observed transient photovoltages from the proteins aldolase,
chymotrypsinogen A and ribonuclease A. They have also attributed the

photoresponse to photoionization of the aromatic amino acids.

PHOTOELECTRIC SPECTROSCOPY 0F CHEMICALLY MODIFIED PM RECONSTITUTED ON
BLMS

(a) Background:

 

The purple membrane of the halophilic bacterium H. halobium is a

 

specialized section of the plasma membrane which developes when the
,cell is growing under low 02 conditions. The PM fragments, which

occur in crystalline arrays, contain a single protein termed
bacteriorhodopsin (bR) because of its resemblance to the visual pigment
of animals (Oesterhelt and Stoeckenius, I971). The protein bR contains
retinal bound to a lysine (Oesterhelt gt 31., 1971) residue in the form
of a Schiff base and exhibits an absorption maximum at 560 nm. Upon
absorption of light, bR undergoes a cyclic photoreaction (Lozier,
Bogomolni and Stoeckenius, 1975) accompanied by the translocation of
protons across the membrane. For a more detailed review on the
structure and function of bR and the PM, the reader is referred to a
review article by Stoeckenius, Lozier and Bogomolni, (1978).

As mentioned previously, several membrane models containing PM
incorporated into different ways have been studied by many
investigators (for a review see Schreckenbach, 1979). In those model
systems photovoltages and photocurrents whose action spectrum (in the
visible region) follows the absorption Spectrum of the chromophore were
obtained. The photoelectric effects are attributed to the proton

pumping activity of bR. In the experiments described in Appendix A,

67

the measurement of the action Spectrum from a PM-BLM model membrane was
investigated in the visible and uv regions. The spectrum not only .
exhibited the usual retinal chromophore peak at 560 nm but also showed
a peak at 280 nm. The peak is attributed to the protein part, however
the nature of the photoeffects elicited by the uv illumination was not
known. The experiments on this section were designed to try to answer
this question. More Specifically, if the photoelectric response is due
to charge separation in the absorbing species itself (as in the case of
chymotrypsin) or comes about through other mechanisms, such as energy
transfer. With this in mind, it was decided to do experiments using
chemically modified and/or bleached purple membranes.

(b) PM modified with N—bromosuccinimide (NBS):

 

N53 is a reagent whnch is known to react with tryptophan, tyrosine
and cysteine (Spande and Witkop, 1967). For these studies however, the
reaction with cysteine is not important since bR does not contain this
amino acid (Bridgen and Walter, 1976). The reaction with tyrosine and
tryptophan is pH dependent, at low pH NBS reacts exclusively with
tryptophan and at neutral or alkaline pH it reacts mainly with
tryptophan, but also with tyrosine (Spande gt al., 1967).

When PM was treated with NBS, it was noted that the purple color
began to fade as the reaction progressed. This loss of purple color by
the PM was also dependent on the concentration of NBS with more color
fading the higher the NBS concentration. These results are consistent
with those obtained by Konishi gt 31. (1977) and indicate that the
interaction of the retinal chromophore with the rest of the protein is

being disturbed.

68

The absorption spectrum of PM treated with NBS at pH 5 (see Figure
3.4) exhibits not only a decrease in absorbance of 280 nm (tryptophan
and tyrosine absorption region) but it also Shows a decrease in
absorbance in the chromophore band at 560 nm.

Liposomes containing PM chemically modified with NBS were added on
one side of a BLM in the presence of Ca2+. There BLMS exhibited
photovoltages when illuminated with visible and uv light, whose
polarity and the electrical characteristics were similar to those
exhibited by BLM with unmodified PM. The only significant effect was
in terms of the magnitude of the photovoltage which decreased the
higher the NBS concentration used. After measuring the action spectrum
of the photoresponse for each preparation, it was observed that both
action spectra, for modified and unmodified PM, were Similar in shape
(see Figure 4.1) both exhibiting a structure at 280 nm. In the case of
modified PM, however, a decrease in the height of the 280 nm peak
relative to the 560 nm peak was observed (see Figures 4.1 and 4.4A).

As shown by Figure 4.4A, the decrease eventually reached a minimum
value so that the relative height of the 280 nm to the 560 nm peak
reached a constant value. Also, the magnitude of the photovoltage at
all wavelengths decreased with increased NBS concentration. The
results show that tryptophan is involved in at least part of the
uv-elicited photovoltages and also suggest that the intact chromophore
is required for their appearance. The results also suggest that intact
tryptophan residues are required for the proton pumping activity of bR,
since modification of these amino acids with NBS reduces the magnitude
of the photovoltage. This last observation is in agreement with those

of Konishi and Packer (1977). The results also give support to the

.5: com um vw~PFmecoc mew:
mm>czo use .oe .m w>csu mom .v w>c=u moH .m m>c=u me .~ m>g=o mo .H m>e=u "mew
mowumc cm—os one .mn on mmz we owpmg Lm—oE ucmgmmwwv m mpcmmmgamc m>e=o comm

.mmz saw: um_epuoa mcmcneme
m_acsa mcwcwwucou mm_u_mm> cur: vows» 24m soc» mguumqm cowuum owoupo>oposm . H.¢ mesmwu

7O

 

 

 

 

F. N M 1' m
I l L 4 l l
O O G 6 O
m V N F‘

(A‘")

afieuonowqd among

700

400

300

("I")

)t

71

idea that the bathocromic (red) shift in bR is due to some solvent
effect caused by protein, i.e., various interactions with amino acid
residues in the immediate vicinity of the chromOphore.

(c) Bleaching with hydroxylamine (NHZOH):

 

The reaction of NHZOH with bacteriorhodopsin in the presence of
light breaks the Schiff base bond of the retinal, forming a retinal
oxime. Once this covalent bond is broken, the interaction of retinal
Schiff base with the protein, which is responsible for its absorbance
at 560 nm diminishes and there is an equivalent increase in the retinal
oxime peak at 355 nm (see Figure 3.5).

Liposomes containing pM bleached for different periods of time
were added to one side of planar BLM. When the BLM was illuminated, a
photoresponse could be detected which had Similar polarity and
characteristics as the one from unbleached PM. However, as in the case
of PM treated with NBS, the magnitude of photoresponse decreased with
increased bleaching of the PM. This result is not unexpected Since
bleaching of the PM should destroy its proton pumping activity and
therefore destroy the photoresponse. The uv elicited photoresponse,
however, need not decrease unless the intact chromophore is necessary
for it to be generated. That this is actually the case was then found
out when the action spectrum of the photoresponse for partially
bleached PM was measured (see Figure 4.2). The action Spectrum exhibi-
ted the same decrease in photovoltage at all wavelengths, including the
280 nm band. These results suggest that the uv elicited photoresponse
is due to energy transfer presumably from tryptophan and tyrosine
residues to the retinal chromophore. This interpretation is consistent

with the results obtained by chemical modification with NBS.

72

 

 

.mesee wN tee eeeeeee=__e o--o--o wee
.cowuu:PE:~_v o: ¢||¢||o .oewp mo ucaoso ucmcwmwwu a com cowuchEsppw mucmmmeamg

m>c=u comm .u;m_P op mcamoaxo vcm :o :z a:_m= vasomopn coon mm; 2a one

.Iomzz spy: umsuompn ozoLnEwE
m_qc=q mcwcwwucoo mo_owmm> ;u_3 cmmzw 24m Eocw acuumam copuum owmgpo>opoca . N.¢ mesmwu

73

 

 

 

 

(A‘")

Photovoltago (mV)
V' M N F.
I I I W
.5
I
,
I
J G
I .
I
z 3 ’
a ’ ’ ’
a
I ’6
6
R \
“8.1
\
\ \@\
\
\.
‘ b
\
\
7 Q
I
0 b
|
|
it Q
\
u Q
I
I
on
I
’ v I
av " ‘ ’
00’
n L 1 1 1
g C G G O
I” Q N N "‘

ofinuoaoaoqd

700

400

300

(nm)

A

74

(d) Bleaching with NaBHA:

 

Another series of experiments similar to the ones performed with
NHZOH and NBS were done using NaBH4. The reaction of NaBH4 with
PM, which only takes place under illumination, results in reduction of
the Schiff base with concomitant discoloring of the PM (Oesterhelt gt
.gl., 1976). The discolored membrane exhibits the typical retinyl
protein fluorescence and does not recombine with retinal to yield
bacteriorhodopsin (Oesterhelt gt 31., 1974). The resulting reduced
form of bacteriorhodopsin shows absorption spectra with maxima at 360
nm and 280 nm. After a series of spectroscopic experiments Peters, gt
El: (1976) concluded that the NaBH4 reaction in the light apparently
modifies the covalent linkage between chromophore and protein while
preserving existing or possibly establishing new non-covalent
interaction(s).

Again, different liposome preparations were added to one Side of a
preformed BLM. Each liposome preparation contained PM bleached to a
different extent by exposing it to NaBH4 and light for different
periods of time. As in the other two cases, the BLM exhibited
photovoltages which were also Similar in characteristics to that of
unbleached PM. Another similarity was that the magnitude of the
photovoltage decreased with increased bleaching of the PM. When the
action spectrum was measured, however, a peak at around 360 nm together
with the 560 nm and the 280 nm, was observed (see Figure 4.3). This
peak at 360 nm, which corresponds to the reduced form of the
bacteriorhodOpsin as reported by Peters gt 21- (1976), was also
observed in the absorption Spectrum (see Figure 3.6). When the action

spectra for several BLM-liposome preparations containing different

75

 

.5: com um mo~w_meLo: ocoz mo>c=o
one .:_E om “m o>c=o ”cwe mm ”m o>g=o ”are mH "m o>L=o “cwe m "N o>c=o “cwe
o ”H o>e=u .mowtoa we?» ococowwwm m to» ogmwp op ocamoaxo mucomocqoc o>c=o comm

.oemw_ mem ezmmz sow; monomepe meeeeEee
opacsg mcwc_mocoo mo—o_mo> ;u_3 moms» zom some mcuooam copuom owmu—o>opo;a i m.¢ ocsmwo

76

mo

 

 

4

l

 

" 500

(I'm)

 

 

10-

(A'“)

l
‘D V

afienonozoqd

N

angmau

400

300

X

77

amounts of bleached PM were measured and compared, the following
results were obtained. The peak at 360 nm increased relative to the
560 nm peak from a small hump, for slightly bleached PM, to a maximum
height and did not increase more even if the PM membrane was bleached
more as could be seen from its absorption Spectrum (see also Figure
4.48). At the same time and similar to the two other cases previously
mentioned (N85 and NHZOH treatments), the magnitude of the
photovoltage at all wavelengths decreases with increased bleaching.

These results can again be explained in terms of energy transfer.
For the 360 nm band, however, the energy transfer is inter-molecular,
from one chromophore to another in the trimer of bR molecules in the
PM. For the 280 nm band, as in the other two cases, it should be
intra—molecular energy transfer from the aromatic amino acids of the
protein backbone to the retinal chromophore.

(e) Summary:

When the action spectrum of the photoresponse of a BLM containing

PM from H. halobium is measured, a 280 nm band is observed together

 

with the expected broad band peaking at 560 nm. Using bleaching and
chemical modification of the purple membrane it was found that the uv
elicited photoresponse was due to energy transfer from the aromatic
amino acids of the bacterirhodopsin protein to the retinal chromophore,
most probably from the four tryptophan residues close to the retinal
moieity. This energy transfer is intra-molecular. There is not much
overlap between the fluorescence emission Spectrum of the aromatic
amino acids (see Chen, 1967) and the absorption Spectrum of the
chromophore. However, the close proximity between the chromophore and

the four tryptOphan residues (see Packer and Konishi, 1978) suggests

78

Figure 4.4 - A) Change in the height of the 280 nm peak relative to

B)

that of the 460 nm peak as a function of NBS
concentration. The change is expressed as the ratio of
photvoltage at 560 nm to photovoltage at 280 nm.

Change in the height of the 360 nm and 280 nm peaks
relative to the 560 nm peak as a function of
illumination time in the presence of NaBH4. The

changes are expressed as ratios of photovoltages at the
given wavelengths.

79

 

 

 

 

 

 

 

 

 

 

A
.35 560/280
:3 e
a:
o
o
o
=
o
>
o
t
o
.1:
‘ 7 I L l l ’
o 10 20 30. 4O
[NBSJ/Ebll]
4.0h
2 3.0 -
b
o
a:
8
u 2 0 "' '
: g - 560/360 ..
a N U—
; —-—x ~X- 560/280
h
o .
f 1-0 . ;_ 1Anzac/280
(L I I n A A A 1 j
0 20 4O 60 80

1 (min)

80

that the energy transfer may be by exciton migration. It was also
found that NaBH4-treated PM containing reduced bR protein, could also
generate photovoltages when reconstituted in BLM model membranes. The
action spectrum of the photovoltages exhibited a new band at «:360 nm
which corresponds to the absorption by the reduced form of bR. The
appearance of this band was also attributed to energy transfer.
However, in this case the energy transfer is inter-molecular and since
there is considerable overlap between the fluorescence emission
spectrum of the reduced bR (see Peters gt $1., 1976) and the absorption
spectrum of the normal bR, most likely the energy transfer is also by
exciton mechanisms. This is also supported by linear dichroism studies
(King, Bogomolni, Hwang and Stoeckenius, 1977) from which a distance
between chromophores of 18.6 A was estimated. Moreover, studies on
circular dichroism of DR indicate that the interaction of the
chromophore neighboring bR molecules produces exciton coupling between

them (Becher and Ebery, 1976).

CHAPTER 5

PARTIAL RECONSTITUTION OF THE PSI
REACTION CENTER OF PLANTS ON BLM
MODEL MEMBRANES

INTRODUCTION

(a) General:

As mentioned previously, one of the approaches used to study the
mechanism of action of membrane components is that of reconstitution of
the purified component, or of membrane fragments containing it, into
membrane models.

The main purpose of the studies and experiments reported in this
section is to try to reconstitute, as least partially, the P51 reaction
center (RC) of thylakoid lamellae into a planar BLM. It iS hoped that
from the final form of the reconstituted system, it will be
demonstrated, by direct electrical measurements, if generation of a
potential difference across the thylakoid membrane takes place as a
result of absorption of light by the pigments in the RC. The specific
method or technique which will be used to try to achieve
reconstitution, is that of fusion or association of vesicles containing
the P51 RC with planar BLMS. This technique may or may not result in

the eventual incorporation of the RC into the BLM.

81

82

(b) Background:

 

Several membrane components, including: photosynthetic RCs, visual
transduction proteins, ion channels and others have been reconstituted
into BLMS using the "fusion of vesicles" approach (reviewed in Chapter
2). Some results seem to indicate that when the vesicles are
hyperosmotic with respect to the bathing solution and Ca2+ is
present only on the same Side as the vesicles, actual incorporation or
insertion of the vesicular membrane or its components into the planar
BLM, takes place (Miller gt 11., 1976; Latorre 23.21-9 1982). In some
studies, purified membrane components are used while in others,
membrane fragments from physiological systems are used.

In previous studies using purified PSI RCs isolated from pea
chlor0plasts Barsky gt_al. (1976) could not obtain direct measurement
of light-induced electric effects. In their experiments, the
generation of photoelectric effects was observed by means of the anion
probe PCB‘. This was attributed to either lack of association
between the vesicle and the thick (colored) planar membrane or to
inactivation of the RC5 by the solvent present in the membrane. In the
experiments to be reported here, thylakoid membrane fragments will be

used instead of isolated RCS.

RESULTS

(a) Association of TMF-vesicles with planar BLMS:

 

Addition of vesicles containing thylakoid membrane fragments
(TMFS) to one Side of a planar BLM resulted in the generation of
light-induced membrane potentials. The generation of the photovoltage

required the addition of artificial carriers of reducing equivalents

83

(e.g., phenazinemethosulfate, (PMS)) to the same side as the vesicles
or the presence of vitamin K3 (Vit K3), a quinone, in the vesicular
membrane (see Figure 5.1). It also required the presence of Ca2+

(or MgZ+) on the bathing solution. The polarity of the

photovoltage indicated that the side containing the liposomes becomes
negatively charged. Figure 5.1 also Shows that addition of ascorbate
to the same side as the PMS resulted in the decrease or disappearance
of the photovoltage. This may be due to the reduction of PMS by ascor-
bate as evidenced by a change in the color of the suspension from light
yellow to green. The reduced PMS (PMSH) will not be able to function
then as electron acceptor. Addition of such chemical agents to the
other side had no effect on the system, suggesting that the photoactive
species had no access to the other Side of the planar membrane. This
plus the fact that no significant change in the BLM conductance was
observed after addition of the vesicles, was interpreted as an
indication that the TMFS were not becoming incorporated into the planar
BLM. This means that only association or what has been termed partial
fusion of the vesicles with the BLM was taking place.

(b) Kinetics of the associationgprocessz

 

The development of the open-circuit photovoltage as a function of
incubation time was investigated. It was observed that the magnitude
of the photovoltage increased with incubation time, rapidly at the
beginning and then leveled off until a maximum was reached (see Figure
5.2). Assuming that the magnitude of the measured photovoltage is an
indicator of the number of vesicles associated with the BLM the curve
of Figure 5.2 Shows the kinetics of the association process. In order
to see if the maximum value was a function of the concentration of

vesicles in the aqueous phase, the experiment was repeated for

 

84

.2: H xpoomswxocaqm mm: cowoapom mc_;umm ogo co cowomcpcoocoo m

qu one .cowumcmqoca oFoPmo> ocoooa oHuH mo o_omc empoe m um oc__o;o_»m_om;amoza
ozo new: moxwe mm: my uw> .mopowmo> on» mm omwm oEmm ogu co copospom

mcwzumm o5» op oumncoomm :5 m.o mcm mza :2 m.o ”m=o_owmm< .mpomu z Ho.o mcm

Am.“ zav commas Iciocmowcp z Ho.o .Fox 2 H.o mo moumwmcoo ocsust cowumnzocw on»

.zom m mo omwm oco op mommm moFowmo>ioZP x2 cowumcocom ommopo>ouosa uwzocwoucoao . H.m ocsmwo

85

:0

>5 o.
to to

 

do. . do -
. . g ... 1 o

 

 

4-

Mil M! m> ..UJHOUQU “2‘

86

different vesicle concentrations. It was found that approximately the
same maximum value was reached except that at different times (see
Figure 5.2), indicating that the maximum was not dependent on the
vesicle concentration. This also suggests that the association may be
an irreversible process.

To test this possibility EDTA in excess of Ca2+ was added to
the side containing the vesicles before the maximum had been reached.
In this case it was found that the phot0potential did not increase much
more after the EDTA was added (see Figure 5.2). At the same time no
decrease in the photovoltage was observed indicating that the vesicles
did not become dissociated with the BLM. It was concluded therefore,
that the process was irreversible.

The fact that a maximum value is not reached at longer incubation
times could be due to inactivation of the vesicles in the aqueous
solution. If this is the case, then addition of fresh vesicles to the
incubation side could result in a sharp increase in the photopotential.
However, when the experiment was done it was found that only a small
increase took place. This suggests that the maximum value is due to
the limited area of the membrane and the subsequent limited number of
binding sites.

The data from Figure 5.2 was plotted again changing this time the
independent variable from time to liposome concentration (see figure
5.3). From the figure, it appears that the photovoltage varies
linearly with concentration. In another experiment liposomes
containing different amounts of thylakoid membrane fragments (TMFS)
were prepared and incubated on one side of a BLM. Figure 5.4 shows a

plot of the time course of the photovoltage development. The

87

.mo opmo>
om» mcwc_mucoo omwm ogo co ocomoca mm: :5 m.o mo cowomgucoocoo m om mzo .m.~ In
um Iouocwowcp z Ho.o .mpomo 2 No.0 ._ux : H.o mo moomwmcoo cowuspom mewsumm one
.2: cm mm: cowmcoamam opowmo> ocu cw copumeucoocoo m Fcu one .on m.~ mcm Axv
m.H .Amv mo.o ”ocoz Ps\mwawpocamoca as mm mommocaxo .mommm moPo_mo> mo uczoEm on»

.ommca mzoozcm on» op mommm
mopo_mo> mo moczoEm ocoeoww_m Loo ucoeao_o>om ommu~o>oooca oco mo omcsoo o5?» . ~.m ocsmwo

88

Ere

av

2.2.:
an

2.:
cu

 

-

oiioii‘

s <23

0

L

 

:—

m—

cm

afieuonowud

(Alli)

89

 

 

mi
3 .
S .
E
6 .
:3 II
as 4-
co
> .
ct
fl
.2
a. 2'
. - , ‘ j
0 0.5 1,5 2.5

Vesicle Conc. (in mg PL/ml)

Figure 5.3 - Variation of the photovoltage as a function of the aqueous
phase vesicle concentration.

9O

.N.m oczmwu :_ mm mcowowmcoo
Losoo .on z: o.~ mcm Axv E: o.H .A4v z: m.o .on :3 ~.o ”ocm mcowumcpcoocoo
m _:u on» .Pe\oo me m.H .oEmm ogu mw mommo F—m co mommm mopowmo> we uczosm oz»

.m Pgo mo moczoEm
ococowo_m mc_c_mocoo mopowmo> to» ocoanPo>om ommppo>ouoga on» we omczoo o5?» . ¢.m oczmwo

l l

a
“

aseuonoioqd

 

40

30

20

10

(mm

Time

92

photovoltage as a function of the amount of TMF in the liposomes was
also obtained from this data and is shown in Figure 5.5. As before the
magnitude of the photovoltage varies linearly with the amount of TMF in
the liposomes and hence, attached to the BLM. Deviations from
linearity at high concentration of TMF may be expected due to the
limited Size of the vesicles and consequently, the amount of TMF that
becomes incorporated into the liposomes.

(c) Action Spectrum of the,photore§ponse:

 

In order to identify the species responsible for the photoeffect,
the action spectrum was measured. Figure 5.6 Shows the action spectrum
of the photocurrent corrected for constant nunber of incident photons
and also the action spectrum of the PSI reaction center obtained from
Ried (1972). The similarity between both curves provides evidence that
the P51 reaction center is directly involved in the phototransduction
process.

(d) Open-circuit photovoltage:

The time course of the open circuit photovoltage for a typical
membrane is shown in Figure 5.7. The photovoltage increased when
coenzyme 05 (C006 or ubiquinone-BO) or Vit K3 were present in the
vesicular membrane, and decreased sharply when o-phenanthroline was
added to the side containing the vesicles (see Figure 5.7). The figure
shows that the photoresponSe when quinones are present in the vesicular
membrane consists of the superposition of two components, one slow and
one fast. The fast component resembles the photoresponse obtained in
the presence of PMS only.

When the BLM is shunted with an external resistance of the same

order of magnitude as that of the BLM, the photovoltage becomes

93

 

 

; 15 -
5
0
u .
fl
:
a
>
6
«fl
2
a. 5r
. l l l ’
o 1.0 2.0

Chl a Conc. (11M)

Figure 5.5 - Variation of the photovoltage as a function of the amount
of Chl a in the vesicles.

 

94

.mcouosa ocomwocw mo Longs:

ocmomcoo coo mooooccoo mm: ococcaoooosa ogp .5: com um Escuooam cowuom Hm;

on» o» moNPFmELo: mm: unocczooooga one .cowmcoomsm opowmo> on» c. ucomoca mm:
mg ow> mcm cowozpom newcomm oco cw ucomoco mm: :5 m.o we cowumeocoocoo m um mza

.zom on» now: mopmwoommm moPowmo>uozH ocu to» omcoamoc oPLuooFoouoga on» mo
Escoooam co_uom ucocczoouoga mcm ANNmH .movm Eocov um Hma on» yo Eaeuooam cowuo< . m.m ocamwo

95

ewenb m3" .Bumoxa 1o Kouogoma anne'ou

 

 

° 9
3 3 In R
I ' I U
I
a ‘
fl“
\
.Q
7‘.
.

 

 

’/
...C
/’ ‘T.
,0
‘x
‘0
l l 1 L 1 1 1 I 1
=. w «a s. N
wt 9 O G O

(snun omens) wounooaoud

700

500

400

(lllll)

A

Figure 5.7 -

96

Effect of C005, o-phananthroline, CCCP and of
external shunt resistance on the open-circuit
photovoltage.

A)

B)

Effect of quinones (Vit K and C006 ) and
o-phenanthroline on the pgotoelectric response (rf
the TMF-vesicles associated with the BLM. The
quinones were mixed with the phospholipid during
vesicle preparation (molar ratio 1:10 quinone:
phosphatidylcholine). The PMS and o-phenanthroline
were both added to the same side as the vesicles to
a final concentration of 0.5 mM and 2 mM
respectively. The incubation mixture was as in
Figure 5.1 The Chl a concentration in the aqueous
solution was m1 uM.

Effect of shunting the Open- -circuit photovoltage
with an external resistance R and by

additions of CCCP (O. 5 uM). AII other additions,
the incubation mixture and the Chl a concentration
were as in A.

C006
or o-phoncn-
PMS Vit K throline

 

 

PMS . 30 s
Vit K3 ccce '

B. l T T i

 

off off
RBLM= 2 x109 :1

off R5x1 “09 9 “31M: 4 " '08 Q
12

Rex-T > 1012 Q

98

differentiated. Addition of small amounts of the protonophorous
uncoupler carbonylcyanide-m-chlorophenylhydrazone (CCCP), which is
known to increase the permeability of membranes to protons (Le Blanc,
1971), cause a small decrease in the membrane resistance and a decrease
in the amplitude of the photovoltage. However, no significant change
in the shape of the photovoltage time course was observed, with the
exception that if a quinone was also present the rise and decay time of
the Slower component decreased (see Figure 5.8). This observation
suggests that the slower component may be associated with ion (probably
H+) diffusion across the membrane. The time constant of the slow
component was calculated from the time course of the photovoltage shown
in Figure 5.8 and found to be equal to 6 S. This value is in agreement
with the value of 5 5 calculated from the values of BLM resistance

(Rm = 1.2 x 109 n) and capacitance (Cm = 4.1 nF) for that

membrane.

Open-circuit photovoltage measurements as a function of light
intensity exhibited a saturation at high intensities (see Figure 5.9).
The response of the membranes to excitation by short (8 us duration)
light flashes was also investigated. The results are Shown in Figure
5.10. The rise time of the photopotentials was found to be of the
order of 50 us.

It was also found that the open-circuit photovoltage obtained in
the presence of PMS and in the absence of any other exogenous
substance, developed after some time, a Slow component. The slow
component, as in the experiments with quinones, had the same polarity
as the fast component and was superimposed on it (see Figure 5.11).

After doing some experiments it was determined that the development of

99

VII K3
oui ‘5 n""
I, 6»:
h—Z’
(off

Figure 5.8 - Time course of the Open circuit photovoltage in the
presence of Vit K3.

The amount of Vit K3 and the incubation mixture are as
in Figure 5.1

lOO

.N.m ocamwo cw mm ocsuxwe cowumnsocH

.xpwmcoucw osmw— sow: ommupo>ouoza uwzocwoucoao ocu mo :o_umwcm> - m.m ocsmwo

191

:3

£533....
a”

£23... 2»...
3

 

 

 

d

 

 

3

afieuonowud

(Am)

192

,-l._

 

 

 

’
arm 3:1\ I
a.» 3.
..l.
- .
. ill
.=.m 53" n
N :3»

fldmcwm m.Ho . diam nocxmm om azm ocmzundxocda czofio<odflmmm A: 1mmuozmm ao Addcaflzmdfioz 2*”: m:
m um A: acxmnwos «dam: om dfimyw.

arm mxwoz maafinmnmm nzm osmmw om nsm flammx. arm danccmndoa adxflcam zmm mm A:
mdccxm m.H . vzm ac m «damd noanmznamn*o= om 0.0m a: 2mm usmmmaa on firm mmam
mfiam mm «:m <mmdndmm.

133

FIRES

0*" 0‘" in O“
T’ | [ "1F
’ i
5 va °“ off f . f

30 s 0“ 0“

Figure 5.11 - Deve10pment of the 10w component of the photovoltage in
the presence of 0.5 mM PMS. Incubation mixture as in
Figure 5.1.

104

this slow component was dependent on the amount of time the membrane
was illuminated. The rise and decay time of the component was found to
be of the same order as the one observed for the quinones. It is sug-
gested then that the component is ionic in nature, most probably due to
H+ diffusion. This idea is also based on the results of experiments
with sonicated chloroplasts in the presence of PMS by Nelson, Nelson
and Racker (1972). In their experiments it was observed that sonicated
broken chloroplasts were capable of ATP synthesis in the presence of
PMS. The interpretation given was that reduction of PMS was resulting
in the net accumulation of PMSH, an amine, indside the sonicated
chloroplast. The reduced PMS (PMSH) inside was then donating electrons
to P700+ with the concomitant release of H+. The net result of the
PMS-mediated cyclic electron flow being the lowering of the internal pH
affecting the polarity of the membrane and creating the necessary
conditions for cyclic photophosphorylation (Nelson gt_gl., 1972).

(e) Short-circuit photocurrent:

 

The time course of the short-circuit photocurrent for a typical
membrane is shown in Figure 5.12. It consists of a large initial
overshoot when the light is turned on, followed by decay to a smaller
steady-state current and an undershoot when the light is turned off,
followed by a return to the base line. When the conductance of the
bilayer to protons was increased with CCCP the steady-state current
also increased. These results taken together with the fact that no
change in conductance was found for the BLM in the presence of the
vesicles strongly suggest that the closed structure of the vesicles is
maintained intact after fusion. The same model has been previously

proposed for studies involving the fusion of planar membranes with

135

10 M

2 5

off

 

 

Figure 5.12 - Time course of the short circuit photocurrent for
TMF-vesicles associated with a BLM.

PMS present at a concentration of 0.5 mM. The incubation
mixture was as in Figure 5.1.

106

vesicles containing purple membrane from H. halobium (Herrmann gt al.,

 

1978) or bacteriochlorophyll complexes (Barsky gt_al., 1976).

A drawing showing the suspected configuration for the
TMF-containing vesicles fused with a BLM is shown in Figure 5.13.

Assuming this configuration is correct, the polarity of the
measured open-circuit photovoltage indicates that when the light is
turned on, the outside of the vesicles charges negative and the inside
positive. In order to test the validity of the proposed model, the
system will be analyied on the basis of an equivalent circuit similar
to the one used by Herrmann and Rayfield (1978). This is justified by
the similarities in the characteristics of the photoelectric response
between this system and that of PM-vesicles fused with BLMs (Herrmann
.gt_al., 1978).

The equivalent electrical circuit for a single fused vesicle is
shown in Figure 5.14. In the diagram Rv and Cv are the resistance
and capacitance between the right chamber and the interior of the
vesicle. Rf and Cf are the resistance and capacitance between the
left chamber and the interior of the vesicle (fused portion). Rm and
Cm are the resistance and capacitance of the planar BLM. All the
equivalent current generators in one vesicle are designated by I and
Vb represents the potential difference between the interior of the
vesicle and either chamber. The shape of the observed transient
short-circuit photocurrent as shown in Figure 5.12 is in agreement with
this electrical circuit. The photocurrent is made up of two
components, a displacement or capacitative current through Cf and a
leakage current through Rf which starts to appear as the light is

turned on. As the potential Vb reaches its steady-state value, the

107

 

 

d Electrodes P

v me v
m m

U Rv

 

 

 

 

 

 

PBLM

 

 

1

Figure 5.13 - Diagram representing the suspected configuration of the
vesicle-BLM system after association of the two
membraneous structures.

108

 

 

 

 

 

 

 

 

———u {IV
V

c:f c

 

 

Figure 5.14 - Equivalent circuit diagram to represent the vesicle-BLM
associated system as shown in Figure 5.13.

Under short-circuit conditions, the potential difference
between a and b vanishes and If would represent the
short-circuit current.

109

displacement current gradually decreases until only the steady-state
current is left. When the light is turned off, the membrane begins to
discharge (Vb decreases), this generates a displacement current of
opposite sign. However, if the displacement current is larger than
the leakage current, the short-circuit current reverses sign and
gradually decays to zero.

In general, the current I depends on the voltage Vb and as a
first approximation it may be presented as a linear function of Vb
(Herrmann and Rayfield, 1978):

I (Vt? t, J) - Io (t,J) (1-Vb/Vc), (1)
where Io, which depends on the light intensity J, is the current for
Vb=0 and Vc is a constant. Under short-circuit conditions the
total voltage across the membrane vanishes and the part of the planar
membrane with no vesicles attached may be omitted in the circuit
analysis. From the circuit it follows that:

I=Vb/R + c dVb/dt (2)
where 1/R=1/Rf + 1/Rv and C=Cf + Cv‘ Substituting for I from
equation (1) into equation (2):

I =Vb (1/R + IOIVC) + dVb/dt (3)

Since the light is turned on at t=0 and off at t=to, the
following values for Io(t) can be set. Io(t)=0 for t<0,

Io(t)=Io for 0<t<tO and Io(t)=0 for t>0. Using these values in
equation (3) and solving the differential equation for Vb(t), the
following equation is obtained:

Vb(t)=IoR' (I-e't/T')
where 1/R'=1 R + Io/Vc and T'=R'C. Since the measured

short-circuit current If flowing through Rf is given by Vb/Rf +

110

CdVb/dt, the equation for If is given by:
If = (10 R'/Rf) (1-e't/T' ) + (cf 10 R'/T') e't/T'. (5)
Equation (5) can be written as

If = 1“’+ (1° - I”) e't/T' (6)
where I00 = IOR'lRf represents the steady-state current and
IO=CfR'IO/I' = Ion/C is the short-circuit current at t=0.

Now, from the definitions of I” and R'

1/1” = Rf/IO R' = Rf/R 10 + Rf/VC. (7)

The current I0 is due to electron transfer through the PSI
reaction center and subsequent proton translocation across the
vesicular membrane and should therefore depend on the light intensity
J. Figure 5.14 shows a plot of 1/I0° versus I/J. Since a straight
line is obtained it follows that:

I” = ISJ/ (J + Jl/z) (8)
where I5 is a constant equal to the saturation current and 01/2 is
another constant equal to the half-saturation intensity. This equation
follows from equation (7) if 1/10 is proportional to the inverse of
the light intensity 1/J. A plot of l/IO versus 1/J is shown in
Figure 5.15 and it does show that a linear relationship exists between
the inverse of both quantities. In other words:

l/Io = K1 + K2/J (9a)
where K1 and K2 are constants. This equation can also be written
as:

10 = I: J/(J + J (9b)

1/2)
where, as before, 13 and JI/Z are both constants equal to the

saturation current and the half-saturation intensity respectively.

111

.NH.m mesmwm cw mm mczuxwe cowumnzocH

.=o_pmcws=p_w
co Xu_mcmucw mzu ecu ucwcczo uwsucwuuucogm mg» cmmzpmn a_;mcowumpmm . mH.m mczmwd

112

#6

Au Eerie.
«.9

_.\n
«6

"6

 

 

4"

 

o.\~

j

 

aé

9m

c.~

a."

U!

(1y Iton

113

Equation (9b) indicates that at low light intensities I0 is
proportional to J. From this and the definition of 1' it follows that
1/1' should vary linearly with J at low light intensities. At high
light intensities the light generated photocurrent Io should saturate
since there should be a limited number of light-driven reaction centers
per unit area in the membrane. A plot of 1/1' versus J showing
linearity at low values of J and saturation at high light intensities
is shown in Figure 5.16.

It was mentioned previously that the protonOphorus uncoupler CCCP
caused an increase in the measured steady-state current by increasing
the permeability of the membrane to protons. It has been found that
the resistance of a bilayer varies with the concentration of CCCP
according to the relation I/R = 1/Ro + ac where Ro is the
resistance in the absence of CCCP, a is a constant dependent on the
geometry and c is the molar CCCP concentration (LeBlanc, 1971). From
the relation it follows that at low CCCP concentrations Rf/R is
independent of CCCP concentration and that Rf varies inversely with
c. Putting this information on equation (7) it follows that 1/Ico
should vary linearly with 1/c. A plot of 1/100 versus I/c showing the
linear dependency is shown in Figure 5.17.

The results mentioned in this section support the idea that under
the conditions the above studies were done, the closed structure of the
vesicles is maintained after fusion of the two membrane takes place.
They also show by direct electrical measurements, that the TMFs
reconstituted in such a way are capable of generating photovoltages and
photocurrents which are directly related with the absorption of light

by the PSI reaction center present in the TMF.

114

.NH.m mcamwd cw mm mcsuxwe cowumnzocH

.xp.m=mpcp ugmw_
zuwz A. v pcmccso uwzucwuupgocm ecu Low ucmumcoo ms_u vacammma any mo cowumwcm> I o~.m mesa?»

115

cu

€5.35

3

n

a."

9N

9m

 

i=4

,l/I

(1.5)

116

.NH.m mczmwu cw mm mcspxwe cowumnsucm

.cowpacucmucou auuu saw; ucmcczoogosa uwsucwu-acosm co cowpmwcm> . ma.m mesmvd

117

e—

.—
r.: V

n .503 \u
e

 

I ‘

It

 

col/I

(Pv 01on

118

(f) Incorporation of RCs from thylakoid membranes intogplanar

lfigflsz

In this set of studies the successful incorporation of reaction
centers from plants into planar membranes was pursued once again by
means of fusion of vesicles containing TMFS with BLMS. This time,
however, a procedure developed by other investigators for this purpose
was used (already reviewed in Chapter 2; see also Miller gt 31., 1976;
Zimmerberg, gt 21., 1980). It includes the use of negatively charged
lipids, divalent ions and more importantly, the establishment of
osmotic gradients across the membranes.

Vesicles containing TMFs and hyperosmotic with respect to the
bathing solution were added to one side of a BLM. The conductance of
the membrane was monitored continuously while applying a small external
potential across the membrane. A short time after the liposomes were
added sharp changes in conductance consisting in transient jumps of
unequal duration were observed (see Figure 5.18). The conductance
spikes were superimposed on a continuous step-wise increase in the
background conductance of the BLM. It was found that the membranes
became more unstable after the conductance changes started to appear
and broke when the conductance became too high. To prevent this from
happening and in order to be able to test the photoactivity of the
BLMs, EDTA in excess of Ca2+ was added before the conductance
increase had reached approximately two orders of magnitude. (The
membranes lasted much longer in the absence of the vesicles.) These
results indicate that incorporation of too much natural membrane has a

deleterious effect on the BLM stability.

119

EIIIAI

 

 

l T

Figure 5.18 - Increase in the BLM conductance in the presence of
hyperosmotic TMF-vesicles.

A potential difference of 50 mV was applied across the
BLM. The up-ward arrow indicates the time of vesicle
addition. The amount of vesicles added in ug PL/ml was
about 65. The Chl a concentration in the aqueous phase
was 1 pH. The incubation mixture consisted of 0.2 M KCl,
0.01 M Tricine-OH (pH 7.5) and 5 mM CaClz present on

the same side as the vesicles.

120

At this point, a small addition of PMS to the same side as the
vesicles resulted in the generation of small photopotentials (up to 1
mv maximum). The photovoltage could be observed almost immediately
after adding the PMS. The polarity indicated that the side containing
the liposomes became negatively charged with respect to the other side.
The response could not be observed in the absence of PMS and could be
slightly increased by adding ascorbic acid to the opposite side. In
this case, however, the photovoltages were transient, exhibiting a fast
rise when the light is turned on with a concomitant slower decay to a
steady-state value, followed by decay with a small undershoot when the
light is turned off and a return to the base line (see Figure 5.19).
This shape is reminescent of the photovoltages measured when the
membrane is shunted with an external resistance (see Figure 5.1). It
was also observed that the magnitude of the photovoltage decreased. At
membrane resistance values below 105 n-cm2 no photopotential could
be measured. A small photovoltage was also observed when the
conductance was low since only a small amount of TMFs had been
incorporated into the BLM. (The above observations indicate some of
the difficulties encountered in this set of experiments.) It was also
found that if the PMS was added to the side opposite to where the
liposomes had been added, a very small photovoltage of Opposite
polarity to the one observed before could be measured.

The photovoltage could be generated when the membranes were
illuminated with red (m 675 nm) and blue (m 450 nm) light but not with
green (m 550 nm) light. These observations suggest that the
photosynthetic pigments were involved at least in the initial light
absorption process and partly responsible for the photopotential

generation.

CCCP
ascorbate 0"
(opposfl. o-phonon-
PMS side) throlino
l on I on 1 ML.
+l ' i l , l

l 1 01f

1 nnhn

Figure 5.19 - Open-circuit photovoltage for a BLM with incorporated
TMFS.

Time course of the open-circuit photovoltage after
addition of hyperosmotic vesicles to one side of a BLM
and after an increase in conductance had already been

ob erved. The membrane resistance decreased from 2 x

10 a to 108 n. The final concentration of PMS and
ascorbate after addition were 0.5 and 0.3 mM
respectively. The interior of the vesicle contained 0.25
M sucrose. The incubation mixture for the BLM consisted
of 0.1 M KCl, 0.01 M Tricine-OH (pH 7.5) and 5 mM CaCl2
added only to the same side as the vesicles. After the
resistance had reached that value, 10 mM EDTA was added
to st0p the fusion process. The polarity of the observed
photovoltage indicates that the side containing the
vesicles and PMS becomes negatively charged.

122

Attempts to measure the short-circuit photocurrent were especially
difficult, mostly due to its expected low magnitude coupled with the
large and noisy background (dark) current. Also attempts to measure
the initial rise time of the photovoltage using short-duration (8 us)
light flashes failed. This was mostly due to the small magnitude of
the photovoltage together with the relatively low light intensity of
the flash.

The results obtained in this set of experiments demonstrate that a
reconstituted system consisting of TMFs incorporated into a planar BLM
is capable of generating photovoltages if suitable electron donors and

acceptors are also present.

DISCUSSION
(a) Association of TMF-vesicles with BLMS:

 

Thylakoid membrane fragments when incorporated into phospholipid
vesicles are capable of generating light-induced electric phenomena.
Moreover, when such vesicles are added to the aqueous solution on one
side of a BLM in the presence of divalent cations (e.g., Ca2+ or
MgZT), interaction between the two membranes takes place. The
interaction may result in association (also referred to as "partial
fusion“) or fusion of the two membranes. The above conclusions are
based on the direct measurement by common electrical methods of
photovoltages and photocurrents across the BLM when the TMF-vesicles
and Ca2+ (or MgZT) are present. The photoelectric responses,
whose action spectra followed very closely the action spectrum of PSI
activity, also required the presence of exogenous substances. These
included PMS in the aqueous environment and/or Vit K3 in the

vesicular membrane.

123

It appears that in the absence of osmotic gradients, the TMFs are
not becoming incorporated into the BLM. This stems from the
observation that the exogenous substances only act when added to the
same side as the vesicles, suggesting that the TMFs do not have access
to the other side of the BLM. Moreover, no changes in the background
conductance of the BLM were observed. The results of the short-circuit
photocurrent measurements, in the absence and presence of CCCP,
demonstrated rather conclusively, that the vesicles were associated
with the BLM while still maintaining their closed structures. The kind
of structure that exists at the region of contact between the two
membranes is not really known and can only be speculated upon. Even
though the association is irreversible, it does not seem to involve
rearrangement or exchange of lipids between the structures. This is
inferred from the fact that such processes should induce changes in the
dark conductance of the BLM (Duzgunes and Ohki, 1980) and no such
changes were observed. Therefore it seems that the region of
association consists of the two Opposed bilayers with both hydrophilic
regions in very close contact with each other. The divalent ions seem
to be responsible for bringing and/or keeping the two together by
shielding the negative surface charges on the surface of both, the BLM
and the vesicle. Diffusion of ions through this contact region from
the interior of the vesicle to the other side of the BLM maintains
electrical contact between the two.

The previous results are similar to the ones obtained with other
membrane fragments and components using the same approach (Barsky gt_
.al., 1976; Herrmann gt 31., 1978).

(b) Thegphotoelectric response:

 

The generation of the photoelectric effects from such

124

reconstituted systems required the presence of PMS and/or Vit K3.

The polarity of the photovoltage indicates that the interior of the
vesicle becomes positively charged. Both of these results are
consistent with the results of experiments by Barsky gt 31. (1976)
using vesicle containing purified PSI RCs. In those experiments no
association of the liposomes with the planar membrane took place and
the photoelectric effects were detected by means of the probe PCB'.
The photoeffects required addition of exogenous substances (eg., PMS)
and resulted in uptake of PCB', indicating that the inside was
becoming positively charged. The polarity seems to indicate that the
role of PMS is that of an oxidizing agent (electron acceptor). This
was confirmed by the observation that addition of ascorbate to the same
side as PMS, reduced dramatically the photoeffect, apparently by
reducing the PMS. This is in contrast with the observation by Barsky
gt_gl. (1976) that addition of ascorbate to the vesicle suspension
resulted in enhanced PCB' uptake. This result in their experiments
is difficult to explain based on the observed polarity of the
photoelectric effects since ascorbate is a non-penetrating electron
donor.

The need for addition of such agents in the experiments here
described is most probably due to the fact that during the isolation of
sub-chloroplast particles many of the compounds of the electron
transport chain are destroyed or lost during the extraction procedure.
In the case of isolation of sub-chloroplast particles by detergent
treatment or sonication of chloroplasts, for example, the 02
evolution capacity of chloroplast is destroyed (Spector and Winget,
1980; Schmidt, Radunz, Koening and Menke, 1978) while most plastocyanin
(PCy), ferredoxin (Fd), and NADPreductase are lost (Hall, 1976).

125

In the studies described here PMS, Vit K3 and ubiquinone-30 were
required for develOpment and/or enhancement of the light-induced
electric effects. Ubiquinone-30 and Vit K3 are both quinones very
similar or related to the natural quinones found on the electron
transport chain and function as electron transport intermediates
(Williams, 1977). PMS (an amine) on the other hand, is an exogenous
substance which is known to interact with the P51 reaction center
(Nelson, Nelson and Racker, 1972). When added to chloroplasts, PMS can
pick up an electron from the reducing side of P51, possibly from the
secondary donor (A,B) (Clayton, 1980), move across the thylakoid
membrane (Soha, Izana and Good, 1970) and donate its electron to
P700+. Since a proton is usually transferred across the thylakoid
membrane the process results in "artificial" cyclic
photophosphorylation and cyclic electron flow Nelson, gt al., 1972).
Both Vit K3 and ubiquinone can act as intermediate electron donors to
PSI (Hall, 1976). The presence of either agent in the vesicular
membrane resulted in the development (in the case of Vit K3 together
with PMS) of the observed photovoltage and photocurrent. These
photoresponses could also be reduced or abolished by agents such as
o-phenanthroline, an inhibitor of electron transport (Williams, 1977)
and the proton carrier CCCP.

The above results together with the observation that the action
spectrum of the photocurrent follows very closely the action spectrum
of PSI is taken as evidence that the reaction center (RC) is directly
responsible for the photoeffects observed in these studies. Also, the
results provide direct evidence in support of the idea that light

absorption and subsequent charge separation in the RC results in the

126

establishment of a potential difference across the thylakoid membrane
of chloroplasts. Moreover, the observation that electron donors acting
inside the vesicle donating electrons to PSI and electron acceptors
picking up electrons from PSI from outside confirms the hypothesis that
the RC spans the thylakoid membrane.

At the same time, however, the relatively long values obtained for
the rise time of the open-circuit photovoltage argues against the idea
that the initial charge separation in the RC results in the generation
of a potential difference in the membrane. This point cannot be
answered in the affirmative by means of electrical measurements, since
the initial charge separation in the P51 RC takes place in 10 ps
(Fenton, Pellin, Govindjee and Kaufmann, 1979). If the potential
difference does not arise as a result of initial charge separation,
however, and takes longer time to develop it could be found out by
electrical measurements.

However, the reason for the relatively long value for the rise
time may have arisen from the configuration of the reconstituted
system. In order to investigate this more thoroughly, the electrical
circuit for the configuration shown in Figure 5.13 will be analyzed.
The circuit is shown in figure 5.20. In the Figure, RV, CV, Rf,

Cf, Rm, Cm have the same meaning as in Figure 5.13. E is the
photopotential generator and r1 is its internal resistance. A
represents the interior of the vesicle, B the side of BLM containing
the vesicles, and D the Opposite side. From the circuit it can be
obtained that VAD’ the photovoltage generated across the vesicular
membrane, will be given by:

vAD = vAD (1._ e't/RC) <10)

127

where: VKB = RE/ri represents the steady-state photopotential,
1/R = 1/r1 + 1/Rv + 1/(Rf+Rm), and c = Cv +

Cme/(Cf+Cm). Assuming the internal resistance r1 is much
smaller than RV, Rf and Rm (see also Drachev gt 31., 1976), the
measured time constant Ip, for the Open-circuit photovoltage
generation will be given by:

Tr '=‘-’ r1- (cv + cmcf/(cm+cf)) (11)
Equation (11) shows that the time constant of the photopotential
generation is different from the expected value Tp = ricv due to
the configuration of the system.

Skulachev and his collaborators reconstituted bacteriochlorophyll
(BChl) RCs in a model system consisting of BChl-vesicles associated
with a lipid-impregnated collodion film. Photovoltages with a rise
time faster than 0.3 us (limited by experimental conditions) were
measured which were attributed to the primary charge separation events
in the RC (Skulachev, 1979). It should be mentioned however that this
matter cannot be resolved from their data since the transfer from P870*
to the primary acceptor I and then to the secondary acceptor Qa (a
quinone) takes place in W 200 ps (Clayton, 1980).

One last point to be discussed related to the configuration of the
system is that of the magnitude of the measured photovoltage. From the
circuit of Figure 5.20 it follows that the measured open-circuit
photovoltage across the BLM is given by:

Van = VADR /(Rm+Rf) (12)
From the suspected nature of the region of association between the two
membranes it is most likely that Rr > Rm. As a consequence, the
measured open-circuit voltage is not higher than 1/2 the photovoltage

generated across the vesicular membrane.

128

.PL mucmpmpog

Foccmucv cm npwz mmpcwm cw .m mocsom mam co mm .H Lopmcmcmm “cmccso meupuco:
on» we cowpwucmmmcamc as» use umuauocwcw Ammv oucmgmwmmg Apcsgmv chcmpxm

cw mo cowuamuxm on» spwz ¢~.m acumen cw use on» op Lu—Pevm m? uwzocwo use

.zgm a saw: umpmpuommm mwpopmm> mo acwpmwmcoo
Empmxm on» mpcmmmcqmc Aqumewxocaam cows: pesocwu Paupcuumpm esp mo uppwsmzum . om.m mczmwu

c

 

 

LV

.1. ..
a: lie “.1

 

 

 

 

 

129

(c) Incorporation of RC from thylakoid membranes into planar

§Lfl§:

The results obtained in this set of experiments are interpreted as
an indication that incorporation of the PSI RC from thylakoid membranes
into planar membranes takes place. This conclusion is based mainly on
two observations. First, the step-wise increase in background BLM
conductance suggesting that ion channels had been incorporated into it.
Second, and more conclusive, the observation that the PSI reaction
center became accessible to non-penetrating agents from both sides of
the planar membrane. Other investigators have also successfully
incorporated proteins and membrane fragments using the same technique
of fusion of hyperosmotic vesicles with planar BLMs (Miller gt al.,
1978; Latorre gt 31., 1982). The process seems to involve first,
association of both bilayers aided by divalent cations (as in previous
fusion studies) followed by mechanical stress (swelling of the vesicle
induced by osmotic gradients across the vesicular membrane).

From the observation that photovoltages were generated in the
presence of PMS and ascorbate by blue and red light (not by green), and
by comparison with the results on partial fusion, it was concluded that
the PSI reaction center was directly involved in the process. The
results are consistent with the previous experiments on TMF-vesicles
associated with BLMS and provide more direct evidence that charge
separation and electron transfer in the PSI RC results in the
generation of a potential difference across the thylakoid membrane.

The results also provide strong evidence that the PSI RC Spans the

thylakoid membrane of plant chloroplasts.

130

The results suggest that there are two populations of
TMF-vesicles, inside-out and inside-in. This was concluded from the
observation that photovoltages of Opposite polarity could be Obtained
depending on the side to which PMS and ascorbate were added. There are
several possible interpretations for this Observation. One
interpretation is that there is a preferred orientation Of the membrane
fragments in the vesicular membrane, as is the case of vesicles

containing PM from H. halobium (Racker and Stoeckenius, 1974). Another

 

possibility may be that since the membrane of thylakoids are asymmetric
(Arntzen, Dilley and Crane, 1969; Coombs and Greenwood, 1976), more
vesicles of one orientation associate with the membrane than those of
the Opposite orientation. Finally, it is also possible that intact
vesicles may be contributing to the photoresponse increasing the
magnitude of the photovoltage when PMS is present on the side
containing the vesicles and staying inactive when PMS is present on the
Opposite side.

One last result to be discussed is that Of the shape of the time
course Of the photopotential. The initial fast rise is attributed, as
before, to the charging Of the membrane due to charge separation in the
RC. The decay Of the photovoltage under illumination is most probably
due to ionic leakage incorporated into the BLM as part of the TMF. The
existence Of such ion conductance pathways is evidenced by the changes
in background conductance Of the BLM as the TMFs become
“incorporated" into the BLM. The time constant of the order of 1 s as
measured from a typical response agrees with the values calculated from
the measured values Of membrane resistance and capacitance. The nature

of the small undershoot with the subsequent return to the base line is

131

not known at this time and more investigations will be needed to
clarify this point.

Finally, the small magnitude of the photoelectric effect in this
set Of experiments is probably due to the small numbers of active
reaction centers incorporated into the BLM. One other possibility is
that after incorporation some of the reaction centers become inactive
due to their interaction with the solvent present in the membrane
(Barsky gt_gl., 1976).

(d) Proposed model for thegphotoelectric response of the two

TMF-BLM-vesicle systems studied.

TO explain the mechanism Of the observed photoelectric response in
both BLM systems studies, a model will be proposed (see Figure 5.22).
It is based on the measurement of light-induced voltages generated
across the TMF-BLMS in the presence Of several suitable electrondonors
and acceptOrs. The nomenclature that will be used for the components
of the electron transport chain in the PSI RCs are the same used in
Chapter 2.

The mechanism is as follows. When the membrane is illuminated the
P700 pigment in the RC becomes excited and donates an electron to the
primary electron acceptor A1 (or X). The extent of this charge
separation is not known and therefore it is not known if this first
step results in the generation Of a potential difference across the
membrane. Subsequent electron transfer to A2 and (A,B) takes place.
From (A,B) the electron is transfered to PMS on one side of the
membrane. At this stage a potential difference (first component) is

already established across the membrane.

132

Figure 5.21 - Diagram of the proposed model to explain the possitfle
sequence of events taking place across the thylakoid

membrane responsible for the Observed photoelectric
response.

133

 

 

 

 

 

134

On the opposite side of the membrane electron donors (e.g., Vit
K3, ubiquinone-30, ascorbate, natural electron donors) reduce the
oxidized P700 (P700+) and protons are released on that side Of the
membrane. In the experiments on partial fusion these protons (H+)
diffusing from the interior Of the vesicles across the planar BLM
contribute to the slower component of the photovoltage. In the
experiments on incorporation Of the TMF into the BLM, the H+ moving
across the TMF-vesicle membrane are responsible for the decrease in the

photovoltage during illumination.

BIBLIOGRAPHY

BIBLIOGRAPHY
Alamuti, N. and Lauger, P. (1970) Biochim. Biophys. Acta 211, 362-364.
Amesz, J. and Duysens, L.N.M. (1977) in Topics in Photosynthesis, Vol.
2: Primary Processes Of Photosynthesis (J. Barber, ed.) pp. 149-185,
Elsevier, Amsterdam.
Amesz, J. and Vrendenberg, W.J. (1966) in Currents in Photosynthesis
Research (J.B. Thomas and J.C. Goedheer, eds.) pp. 75-81, Donker,
Rotterdam.
Andersen, O.X. (1978) Membr. Transp. Biol. 1, 369, 456.
Anderson, J.M. (1975) Biochim. Biophys. Acta 416, 191-235.

Anderson, S.S., Lyle, 1.6. and Paterson, R. (1976) Nature 259,
147-148.

Andreoli, T.E. and Tosteson, D. (1971) J. Gen. Physiol. 57, 526-538.

Antanavage, J., Chien, T.F., Ching, Y.C., Dunlop, C. and Mueller, P.
(1977) Biophys. J. 17, 182a.

Arntzen, C.J. and Briantais, J.M. (1975) in Bioenergetics Of
Photosynthesis (Govindjee, ed.) pp. 51-94, Academic Press, New York.

Arntzen, C.J., Dilley, R.A. and Crane, F.L. (1960) J. Cell. Biol. 43,
16-31.

Arthur, W.E. and Strehler, B.L. (1975) Arch. Biochem. Biophys. 70,
507-526.

Babcock, G.T. and Sauer, K. (1975) Biochim. Biophys. Acta 376,
315-344.

Bamberg, E., Anell, H.-J., Dencher, N.A., Sperling, W. and Stiere, H.
(1979) Biophys. Struct. Mechn. 5, 277-292.

Bangham, A.D. (1973) Adv. Lipid Res. 1, 65-104.

Bangham, A.D., Hill, M.W. and Miller, N.G.A. (1974) Method. Membr.
Biol. 1, 1-68.

Barber, J. and Kraan, G.P.B. (1970) Biochim. Biophys. Acta 194, 49-95.
Barsky, E.L., Dancshazy, Z., Drachev, L.A. Il'ina, M.D., Jasaitis, A.A.

Kondrashin, AA., Samuilov, V.D. and Skulachev, V.P. (1976) J. Biol.
Chem. 215, 7066-7071.

135

136

Bauer, P., Dencher, N.A. and Heyn, M.P. (1976) Biophys. Struct. Mechn.
2, 79-92.

Becher, B. and Ebery, T.G. (1976) Biochim. Biophys. Res. Comm. 69,
1'60

Becker, M.J., Shefner, A.M. and Gross, J.A. (1965) Plant Physiol. 40,
243-250.

Bell, D., Haug, A. and Good, N. (1978) Biochim. Biophys. Acta 504,
446-455.

Bengis, C. and Nelson, N. (1977) J. Biol. Chem. 252, 4564-4569.

Bensasson, R. and Land, E.J. (1973) Biochim. BiOphys. Acta 325,
175-181.

Benz, R., Janko, K. and Lauger, P. (1979) Biochim. Biophys. Acta 551,
236-247.

Bishop, 0.6. (1974) Potochem. Photobiol. 20, 281-299.

Bishop, N.I. (971) Annu. Rev. Biochem. 40, 197-226.

Blok, M.C. and Van Dam, K. (1978) Biochim. Biophys. Acta 507, 48-61.
Boardman, N.K. and Anderson, J.M. (1964) Nature 203, 166-167.

Boyer, P.D. (1974) in Dynamics Of Energy-Transducing Membranes (L.
Ernster, R. Estabrook and E.C. Slater, eds.) pp. 289-301, Elseiver,
Amsterdam.

Bridgen, J. and Walker, I. (1976) Biochemsitry 15, 792-798.

Bulychev, A.A. Andrianov, V.K. Kurella, G.A. and Litvin, F.F. (1972)
Nature 236, 175-176.

Chance, B. and Williams, G.R. (1956) Adv. Enzymol. 17, 65-134.
Cheniae, G.M. and Martin, I.F. (1972) Plant Physiol. 50, 87-94.
Chen, R.F. (1967) Anal. Lett 1(1), 35-42.

Cherry R.J., Hsu, K. and Chapman, D. (1971) Biochim. Biophys. Res.
Comm. 43, 351-358.

Chien, T.F. and Mueller, P. (1976) Fedn. Proc. Fedn. Am. Socs. Exp.
Biol. 35, 1599.

Chizmadzhev, Y.A., Abidor, 1.6., Pastushev, V.F. and Arakelya, V.B.
(1979) Bioelectrochem. Bioenerget. 6, 37-87.

Clayton, R.K. (1980) Photosynthesis: Physical Mechanisms and Chemical
Patterns, Cambridge University Press, New York.

137

Cohen, F.S., Zimmerberg, J. and Finkelstein, A. (1980) J. Gen. Physiol.
771, 251-270.

Cohen, J.A., Moronne, M.M. (1976) J. Supramol. Struct. 5, 409-416.
Coombs, J. and Greenwood, A.D. (1976) in Topics in Photosynthesis, Vol.
1: The Intact Chloroplast (J. Barber, ed.) pp. 1-51, Elsevier,
Amsterdam.

Cone, R.A. (1967) Science 155, 1128-1131.

Crofts, A.R., Wraight, C.A. and Fleischman, D.E. (1971) FEBS Lett. 15,
89.

Dancshazy, Z. and Karvaly, B. (1976) FEBS Lett. 72, 136-138.

Darszon, A. Philipp, M., Zarco, J. and Montal, M. (1978) J. Membrane
Biol. 43, 71—90.

Das, M.L. and Crane, F.L. (1964) Biochemistry 3, 696-700.

deGrooth, B.G., van Grondelle, R., Romijn, J.C. and Pulles, M.P.J.
(1978) Biochim. Biophys. Acta 430, 265-281.

DenHaan, C.A., Gorter deVries, H. and Duysens, L.N.M. (1976) Biochim.
Biophys. Acta 430, 265-281.

Donovan, J.J., Simon, M.I., Draper, R.K. and Montal, M. (1981) Proc.
Nat]. Acado SCio UoSvo 78, 172-1760

Doring, 6., Renger, 0., Vater, J. and Witt, H.T. (1969) Z. Naturforsch.
24b, 1139-1143.

Douce, R., Holtz, R.B. and Benson A.A., (1973) J. Biol. Chem. 248,
7215-7222.

Drachev, L.A., Frolov, V.M., Kaulen, A.D., Kondrashin, A.A., Samuilov,
V.D., Semenov, A. Yu. and Skulachev, V.P. (1976) Biochim. Biophys. Acta
440, 637-66.

Drachev, L.A., Frolov, V.M., Kaulen, A.D. Lieberman, E.A., Ostroumov,
S.A., Plakunova, V.B., Semnov, A. Yu and Skulachev, V.P. (1976a) J.
Biol. Chem. 251, 7059-7065.

Drachev, L.A., Jasaitis, A.A., Koulen, A.D., Kondrashin, H.A.,
Liberman, E.A., Hayrecek, I.B., Ostroumov, S.A., Semenov, A.Y. and
Skulachev, V.P. (1974) Nature 249, 321-323.

Drachev, L.A., Kondrashin, A.A., Samuilov, V.V. and Skulachev, V.P.
(1975) FEBS Lett. 50, 219-222.

Duysens, L.N.M. (1954) Science 120, 353-354.
Duzgunes, N. and Ohki, S. (1981) Biochim. Biophys. Acta 640, 734-747.

138

Evans, M.C.W. (1977) in Topics in Photosynthesis, Vol. 2: Primary
Proceses of Photosynthesis, (J. Barber, ed.) pp. 433-464, Elsevier,
Amsterdam.

Fenton, J.M., Pellim, M.J., Govindjee and Kaufmann, K.J. (1979) FEBS
Lett. 100, 1-4.

Fesenko, E.E. and Lyubarsky, A.L. (1977) Nature 268, 562-563.
Fettiplace, R. and Hayden, D.A. (1980) J. Physiol. Rev. 60, 510-550.
Fleischman, D.E. (1971) Photochem. Photobiol. 14, 277-286.

Gitler, C. and Montal, M. (1972) FEBS Lett. 28, 329-332.

Gorter, E. and Grendel, F. (1925) J. Exptl. Med. 41, 439-443.

Graber, P., Schlodder, E.S. and Witt, H.T. (1977) Biochim. Biophys.
Acta 461, 426-440.

Grossweiner, L.I., Brendzel, A.M. and Blum, A. (1981) Chem. Phys. 57,
147-1550

Gunning, B.E.S. and Steer, M.W. (1975) Ultrastructure and the Biology
of Plant Cells, Arnold, London.

Haggins, W.A. (1972) Annu. Rev. Biophys. Bioeng. 1, 131-158.
Haggins, W.A. and McGaughly, R.E. (1968) Science 159, 213-215.

Hall, D.O. (1976) in Topics in Photosynthesis, Vol. 1: The Intact
ChlorOplast (J. Barber, ed.) pp. 135-170, Elsevier, Amsterdam.

Hanke, W., Eibl, H. and Boheim, G. (1981) BiOphys. Struct. Mech. 7,
131-137.

Herrman, T.R. and Rayfield, G.W. (1978) Biophys. J. 21, 111-125.
Helep-Harrison, J. (1963) Planta 60, 243-260.
Heslop-Harrison, J. (1966) Sci. Prog. (Oxford) 54, 519-541.

Higgins, J., Lopez, J.R. and Tien, H.T. (1979) Bioelectrochem.
Bioenerg. 6, 509-524.

Hill, R. and Bendall, F. (1960) Nature 186, 136-137.

Hind, G. and Olson, J.M. (1968) Ann. Rev. Plant Physiol. 19, 249-282.
Hohl, A.R. and Hepton, A. (1965) J. Ultrastruct. Res. 12, 542-546.
Hong, F.T. and Mauzerall, D. (1972) Nature-New. Biol. 240, 154-155.
Hubbel, W.L. and Bownds, M.D. (1979) Annu. Rev. Neurosci. 2, 17-34.

139

Huebner, J.S. (1978) J. Membrane Biol. 39, 97-132.

Huebner, J.S. Arrieta, R.T. and Millar, 0.8. (1982) Photochem.
Photobiol. 35, 467-471.

Huebner, J.S. and Tien, H.T. (1972) Biochim. BiOphys. Acta 256,
300-306.

Ilani, A. and Berns, 0.5. (1972) J. Membrane Biol. 8, 333-356.
Jackson, J.B. and Crofts, A.R. (1969) FEBS Lett. 4, 185-189.

Jagendorf, A.T. and Uribe, E. (1969) Proc. Natl. Acad. Sci. U.S.A. 55,
170-177.

Jordan P.C. (1980) Biophys. Chem. 12, 1-11.

Joshi, N.B., Lopez, J.R. Tien, H.T., Wang, C.-B. and Liu, Q.-Y. (1982)
J. Photochem. 20, 139-151.

Junge, W. and Witt, Z. (1968) Z. Naturforsch. 23b, 244-254.
Karvaly, B. and Dancshazy, Z. (1977) FEBS Lett. 76, 36-40.
Kayushin, L.P. and Skulachev, V.P. (1974) FEBS Lett. 39, 39-42.
Ke, B. (173) Biochem. BiOphys. Acta 301, 1-33.

Ke, B. and Vernon, L.P. (1967) Biochemistry 6, 2221-2226.
Kendall-Tobias, M. and Crofts, A.R. (1979) Biophys. J. 25, 54a.

King, G.I., Bogomolni, R.A., Hwang, S.B. and Stoeckenius, W. (1977)
Biophys. J. 17, 97.

Kirk, J.T.O. (1971) Annu. Rev. Biochem. 40, 161-196.

Klimov, V.V., Klevanik, A.V, Shuvalov, V.A. and Krashnovsky, A.A.
(1977) FEBS Lett. 82, 183-186.

Kobamoto, N. and Tien, H.T. (1971) Biochim. Biophys. Acta 266, 56-66.
Kok, B. (1956) Biochim. Biophys. Acta 22, 399-401.

Konishi, T. and Paker, L. (1977) FEBS Lett. 79, 369-373.

Kreutz, W. (1970) Adv. Bot. Res. 3, 53-169.

Latorre, R., Vergara, C. and Hidalgo, C. (1982) Proc. Natl. Acad. Sci.
U.S.A. 79, 805-809.

Le Blanc, O.H. (1971) J. Membrane Biol. 4, 227-251.
Levine, R.P. (1969) Annu. Rev. Plant. Physiol. 20, 523-540.

140

Liu, Q.-Y. and Tien, H.T. (1982) Photobiochem. Photobiophys. 4, 73-78.
Lozier, R.H., Bogomolni, R.A. and Stoeckenius, W. (1975) BiOphys. J.
15, 95-962. '

Malkin, R. and Aparicio, P.J. (1975) Biochem. Biophys. Res. Comm. 63,
1157-1160.

Malkin, S. (1977) in Topics in Photosynthesis, Vol. 2: Primary
Processes of Photosynthesis (J. Barber, ed.) pp. 349-431, Elsevier,
Amsterdam.

Mangel, M., Berns, D.A. and Ilani, A. (1975) J. Membrane Biol. 20,
171-180.

Mathis, P. (1977) in Topics in Photosynthesis, Vol. 2: Primary
Processes of Photosynthesis, (J. Barber, ed.) pp. 269-302, Elsevier,
Amsterdam.

McCarty, R.E. (1979) Annu. Rev. Plant Physiol. 30, 79-104.

McKinney (1941) J. Biol. Chem. 140, 315-322.

Michel, J.M. and Michel-Wolwertz, M.R. (1969) in Progress in
Photosynthesis Research, Vol. 1 (H. Metzner, ed.) pp. 115-127, H.
Laupp, Tubingen.

Miles, C.D. and Jagendorf, A.T. (1969) Arch. Biochem. Biophys. 129,
711-719.

Miller, C. (1978) J. Membrane Biol. 40, 1-23.

Miller, C., Arvan, P., Telford, J.N. and Raker, E. (1976) J. Membrane
Biol. 30, 271-282.

Mitchel, P. (1961) Nature 191, 144-148.

Mitchel, P. (1968) Chemiosmotic Coupling and Energy Transduction, Glynn
Research, Bodmin, United Kingdom.

Montal, M. (1974) in Perspectives in Membrane Biology (S. Estrada-O.
and C. Gitler, eds.) pp. 591-622, Academic Press, New York.

Montal, M. (1979) Biochim. Biophys. Acta 559, 231-257.

Montal, M., Daszon, A. and Schindler, H. (1981) Quart, Rev. Biophys.
14, 1-79.

Montal, M. Darszon, A. and Trissl, H.W. (1977) Nature 267, 221-225.

Mueller, P. and Rudin, D.0. (1967) Biochim. Biophys. Res. Comm. 26,
398-404.

Mueller, P., Rudin, D.O., Tien, H.. and Wescott, W.C. (1962) Nature
194, 979-981.

141

Mueller, P., Rudin, D.P., Tien, H.T. and Wescott, W.C. (1963) J. Phys.
Chem. 67, 534-535.

Murakami, S. (1964) J. Electron Micros. 13, 234-236.

Nelson, N., Anholt, R. Lindstrom, J. and Montal, M. (1980) Proc. Nat.
Acad. SCio U.S.A. 77, 3057-30610

Nelson, N., Nelson, H. and Racker, E. (1972) Photochem. Photobiol. 16,
481-489.

Oesterhelt, D. and Schumann, L. (1974) FEBS Lett. 44, 262-265.

Oesterhelt, D. and Stoeckenius, W. (1971) Nature- New Biol. 233,
149-152.

Oesterhelt, D. and Stoeckenius, W. (1974) in Methods in Enzymology, 31,
Biomembranes, part A (S. Fleisher and L. Packer, eds.) pp. 667-678,
Academic Press, New York.

Orbach, E. and Finkelstein, A. (1980) J. Gen. Physiol. 75, 427-436.

Packham, N.K., Packham, C., Mueller, P., Tiede, D.M. and Dutton, P.L.
(1980) FEBS Lett. 110, 101-106.

Pant, H.C. and Rosenberg, B. (1971) Photochem. Photobiol. 14, 1-14.

Pattus, F., Desnuelle, P. and Verger, R. (1978) Biochim. Biophys. Acta
507, 62-70.

Pattus, E., Piorant, M.C.L., Lardunski, C.J., Desnuelle, P. and Verger,
R. (1978) Biochim. BiOphys. Acta 507, 71-82.

Park, R.B. and Sane, P.V. (1971) Annu. Rev. Plant Physiol. 22,
395-4300

Peters, J., Peters R. and Stoeckenius, W. (1975) FEBS Lett. 61,
128-134.

Pulles, M.P.J., Van Gorkom, H.J. and Williams, J.G. (1976) Biochim.
Biophys. Acta 449, 536-540.

Racker, E. and Hindle, P.C. (1974) J. Membrane Biol. 17, 181-188.
Racker, E. and Stoeckenius, W. (1974) J. Biol. Chem. 249, 662-663.

Repke, H., Benczi, A. and Matthies, H. (1980) Acta Biol. Med. Germ. 39,
657-663.

Ried, A. (1972) in Proc 2nd Int. Congr. Photosynth. Res. (G. Forti, M.
Avron and A. Melandri, eds.) pp. 763-772, Junk, The Hague.

Robertson, J.D. (1960) Progr. Biophys. Biophys. Chem. 10, 343-418.

142

Ryrie, I.J. and Blackmoore, P.F. (1976) Arch. Biochem. Biophys. 176,
127-135.

Saphone, S. and Crofts, A.R. (1971) Z. Naturforsch. 326, 617-626.
Schindler, H. (1979) Biochim. Biophys. Acta 555, 316-336.
Schindler, H. (1980) FEBS Lett. 122, 77-79.

Schindler, H. and Quast, U. (1980) Proc. Nat. Acad. Sci. U.S.A. 77,
3052-3056.

Schmidt, O.H., Radunz, A., Koeni , F. and Menke, W. (1978) in
Photosynthetic Oxygen Evolution (H. Metzner, ed.) pp. 91-104, Academic
Press, New York.

Schonfeld, M. Montal, M. and Feher, G. (1979) Proc. Nat. Acad. Sci.
U.S.A. 76, 6351‘63550

Seely, G.R. in Topics in Photosynthesis, Vol. 2.: Primary Processes of
Photosynthesis (J. Barber, ed.) pp. 1-53, Elsevier, Amsterdam.

Shamoo, A.E. and Trivol, W.F. (1980) Curr. TOp. Memb. 14, 57-126.

Shieh, P. and Packer, L. (1976) Biochem. Biophys. Res. Comm. 71,
603-609.

Shuvalov, V.A., Dolan, E. and Ke, B. (1979) Proc. Nat. Acad. Sci.
U.S.A. 76, 770-7730

Singer, S.J. and Nicolson, G.L. (1972) Science 175, 720-731.
Skulachev, V.P. (1979) in Topics in Photosynthesis, vol. 3:
Photosynthesis in Relation to Model Systems (J. Barber, ed.) pp.
175-188, Elsevier, Amsterdam.

Soha, S., Izana, S. and Good, N.E. (1970) Biochim. Biophys. Acta 223,
158-164.

Spande, T.F. and Wintkop, P. (1967) in Methods in Enzymology Vol. XI
(C.H. W. Hirs, ed.) pp. 506-522, Academic Press, New York.

Spector, M. and Winge, D. (1980) Proc. Nat. Acad. Sci. U.S. A. 77,
957-959.

Steineman, A., Stark, G. and Lauger, P. (1972) J. Membrane Biol. 9,
177-194.

Stoeckenius, W., Lozier, R.H. and Bogomolni, R.A. (1976) Biochim.
Biophys. Acta 505, 215-278.

Strehler, B. and Arnold, W. (1951) J. Gen. Physiol. 34, 809-829.

143

Szoka, F. and Paphadjopoulos, D. (1980) Annu. Rev. Biophys. Bioeng. 9,
497-598.

Takagi, M. and Koshimoto, U. (1977) Sci. Rep. (Osaka Univ.) 26, 11-17.

Thomson, W.W. (1974) in Dynamic Aspect of Plant Ultrastructure (A.W.
Robards, ed.) pp. 138-177, McGraw-Hill, London.

Thornber, J.P., Aberte, R.S., Hunter, F.A., Shiozawa, J.A. and Kan,
K.-S. (1976) Brookhaven Symp. Biol. 28, 132-148.

Tien, H.T. (1968) Nature 219, 272-274.
Tien, H.T. (1972) Photochem. Photobiol. 16, 271-290.

Tien, H.T. (1974) Bilayer Lipid Membranes (BLM): Theory and Practice,
M. Dekker, New York.

Tien, H.T. (1979) in TOpics in Photosynthesis, Vol. 3: Photosynthesis
in Relation to Model Systems (J. Barber, ed.) Elsevier, Amsterdam.

Tien, H.T. (1981) in Membranes and Transport (A.N. Martonosi, ed.)
Plenum Publishing, New York.

Tien, H.T. and Verma, S.P. (1970) Nature 227, 1232-1234.

Tredgold, R.H. and Elgamal, M. (1979) Biochim. Biophys. Acta 555,
381-387 0

Trissl, H.W., Darszon, A. and Montal, M. (1977) Nature 266, 655-657.
Trissl, H.W. and Lauger, P. (1972) Biochim. Biophys. Acta 282, 40-54.

Tyrell, D.A., Heath, T.D., Colley, C.M. and Ryamn, B.E. (1976) Biochim.
Biophys. Acta 457, 259-302.

Ullrich, H.M. and Kuh, H. (1972) Biochim. Biophys. Acta 266, 584-596.

Van Best, J.S. and Duysens, L.N.M. (1977) Biochim. Biophys. Acta 459,
187-206.

Van Gorkom, H.J. (1974) Biochim. Biophys. Acta 347, 439-442.

Van, N.T. and Tien, H.T. (1970) J. Phys. Chem. 74, 3559-3568.
Verger, R. and Pattus, F. (1976) Chem. Phys. Lipids 16, 285-291.
Vernon, L.P. and Avrom, M. (1965) Annu. Rev. Biochem. 34, 269-296.

Vodyanoy, V, and Murphy, R.B. (1982) Biochim. Biophys. Acta 687,
189-194.

Waldbillig, R.C. and Szabo, G. (1979) Biochim. Biophys. Acta 687,
189-194.

144

Wang, C.-B., Tien, H.T., LOpez, J.R., Liu, O.-Y., Joshi, N.B. and Hu,
Q.-Y. (1982) Photobiochemistry and Photobiophysics 4, 177-184.

Weier, T.E. Engelbrecht, A.H.P., Harrison, A. and Risley, E.B. (1965)
J. Ultrastruct. Res. 13, 92-111.

Weier, T.E. and Thomason, W.W. (1962) J. Cell Biol. 13, 89-108.

Weier, T.E., Stocking, C.R., Thomson, W.W. and Drever, H. (1963) J.
Ulatrastruct. Res. 8, 122-143.

Wikstrom, M. and Krab, K. (1979) Biochim. Biophys. Acta 549, 177-222.
Williams, W.P. (1977) in Topics in Photosynthesis, Vol. 2: Primary
Proceses Of Photosynthesis (J. Barber, ed.) pp.99-147, Elsevier,
Amsterdam.

Winget, G.D., Kanner, N. and Racker, E. (1977) Biochim. Biophys. Acta
466, 315-324.

Witt, H.T. and Zickler, A. (1973) FEBS Lett. 37,307-310.

Wolff, Ch., Buchwald, H.E., Ruppel, H., Witt, K. and Witt, H.T. (1969)
Z. Naturforsch. 24b, 1038-1041.

Yoshida, M., Sone, N., Hirata, H., Kagawa, Y., Takeuchi, Y. and Ohno,
K. (1975) Biochem. Biophys. Res. Comm. 67, 1295-1300.

Zimmerberg, J., Cohen, F.S. and Finkelstein, A. (1980) J. Gen. Physiol.
75, 241-250.

PLEASE NOTE:

NO pages in appendices are missing,
per author.

UNIVERSITY MICROFILMS INTERNATIONAL

APPENDIX A

APPENDIX 433

Biochimica et Biophysica Acta, 597 (1980) 433—444
© Elsevier/North-Holland Biomedical Press

BBA 78698

PHOTOELECTROSPECTROMETRY OF BILAYER LIPID MEMBRANES

JOSE R. LOPEZ and H. TI TIEN
Department of Biophysics, Michigan State University, East Lansing, MI 48824 (U.S.A.)
(Received July 18th, 1979)

Key words: Asymmetry; Membrane structure; Photoelectrospectrometry; (Lipid bilayer)

Summary

Three different bilayer lipid membrane systems were studied under visible
and ultraviolet illumination. The first system consisted of a bilayer lipid mem-
brane formed with a mixture of phospholipids and cholesterol, to one side of
which purple membrane fragments from Halobacterium halobium were added.
The second system consisted of a membrane formed from spinach chloroplast
extract. When either of these membrane systems was illuminated with ultra-
violet and visible radiation, photopotentials were observed and photoelectric
action spectra were recorded (the technique is termed photoelectrospectrom- ,
etry). Each spectrum had a definite structure which was characteristic of each
of the modified membranes. The third system studied consisted of an otherwise
photoinactive membrane formed with a mixture of phospholipids and choles-
terol, to one side Of which chymotrypsin was added. When the membrane was
illuminated with visible light no photoresponse was observed. On the other
hand, a photopotential which increased with incubation time was observed
when the membrane was illuminated with ultraviolet light. Since, in our sys-
tems, the photoresponses have been observed to be due to certain species
incorporated into the membrane, it appears that photoelectrospectrometry is a
useful tool for studying lipid-protein interactions, constituent organization and
energy transfer in membranes.

 

Introduction

Pigmented membranes in living cells serve one of the two vital functions,
energy transduction and signal detection. In the case of the former, the thyla-
koid membrane of chloroplasts is the best example, whereas the sac membrane
of the rod outer segments in the vertebrate eye typifies the latter. The purple
membrane of Halobacterium halobium appears to be capable of either function

145

434

in response to environmental conditions [1]. Since biological membranes are a
complex assembly of lipids, proteins and other constituents such as pigments
and are difficult to study at the molecular level, model membranes have instead
been resorted to by numerous investigators [2,3]. Of the many model mem-
branes, the artificial bilayer lipid membrane of planar configuration is particu-
larly pertinent as a model because its organization is similar to that of the lipid
bilayer believed to exist in biological membranes [4].

In recent years, methods have been developed for studying pigmented
bilayer lipid membranes separating two aqueous solutions [4,5]. Photoactive
compounds such as chlorophylls [6—10], retinals [11-13] bacteriorhodopsin
[14—16] and a variety of dyes [4,17—19] have been incorporated into these
model membranes. To characterize this bilayer lipid membrane system, elegant
spectroscopic techniques have been developed [4,6,8], which provide informa-
tion on the composition and organization of constituent molecules in the
bilayer lipid membrane. In this paper we report the study of bilayer lipid mem-
branes in the ultraviolet and visible region, thereby extending the usefulness of
membrane photoelectrospectrometry [4]: in particular, photoeffects from
bilayer lipid membranes made of chloroplast extract, containing purple mem-
brane and also bilayer lipid membranes to one side of which chymotrypsin has
been added, when irradiated with ultraviolet light. In all three cases, open
circuit photovoltage action spectra which followed the absorption spectrum of
the modifier were obtained.

Materials and Methods

Chlorophyll-containing bilayer lipid membranes. A bilayer lipid membrane
solution containing chlorophyll pigments in butanol/n-octane (1 : 1) was
obtained from fresh spinach leaves by a procedure described elsewhere [4]. The
aqueous solution consisted of 0.1 M acetate buffer at pH 5.

Bacteriorhodopsin-containing bilayer lipid membranes. The membrane-
forming solution for the bacteriorhodopsin experiments consisted of either a
1 : 1.7 : 3.1 : 1.1 mixture of phosphatidylserine/phosphatidylcholine/phos-
phatidylethanolamine/cholesterol in n-octane to a final concentration of 6.9%
(w/w) or a 3.5% (w/w) H. halobium: phospholipids mixture in n-decane. The
aqueous solution for these experiments consisted of 75 mM KC1+ 25 mM
CaCl, (unbuffered, pH - 6.8).

Phospholipid vesicles containing purple membrane fragments were prepared
from egg lecithin following a procedure described by Blok et al. [20]. After the
membranes were formed, vesicles containing purple membrane were added to
one side of the bilayer lipid membrane.

Bilayer lipid membranes with chymotrypsin in the aqueous solution. The
membrane-forming solution consisted of either a 1 : 1.7 : 3.1 : 1.1 mixture of
phosphatidylserine/phosphatidylcholine/phosphatidylethanolamineIcholesterol
in n-octane to a final concentration of 6.9% (w/w) or a 1 : 1 mixture of phos-
phatidylcholine/oxidized cholesterol in n-octane to final concentration of 2%
(w/w).

The membranes were formed in 1 - 10“, 1 - 10"3 and 1 - 10'1 M KCl. After

435

the membranes had thinned, 200 pl of a 5% chymotrypsin solution in double-
distilled water was added to one side of the membrane (vol. 8 ml).

Methods

The technique used in these studies is essentially the same as that described
previously [4]. Bilayer lipid membranes were formed in a 1-mm aperture of a
Teflon cup separating two aqueous solutions. The cup was placed in a plexiglass
chamber With a quartz window. Electrical contact between the two aqueous
solutions was made with a pair of calomel electrodies via salt bridges. The
photopotentials were measured with a 610BR Keithley electrometer connected
to a VOM 6 Bausch and Lomb chart recorder. Closed-circuit measurements
were performed by applying a potential through an external resistance 10—20
times larger than the dark resistance of the membrane. The photovoltage action
spectra in the ultraviolet and visible regions were obtained by illuminating the
membranes with light from a 1000 W Xenon lamp after passing through a 250
mm Bausch and Lamb grating monochromator equipped with a 600 grooves/
mm diffraction grating. The light from the lamp was focused onto the entrance
slit of the monochromator by means of a quartz lens. The exciting radiation
from the monochromator was focused on the 1 mm hole in the Teflon cup by
means of a second quartz lens. The entrance and exit slits of the monochroma-
tor were both set at 3.0 mm. The curves were corrected for constant incident
illumination with a Scientific Instrument Kettering radiant power meter. The
absorption spectra were measured with a DB Beckman spectrophotometer. The
photoresponses from the membranes were all recorded once the membranes
had reached the ‘black’ stage and their resistance was constant. All of the
experiments reported here were carried out at 22 t 1°C.

Results

Chlorophyll-containing membrane

After the membrane had thinned to the black state and its dark resistance
has reached a steady value of approx. 1 - 108 52, it was irradiated with light of
300 nm wavelength. A small photopotential (0.5-2.0 mV) negative on the
illuminated side was observed under open-circuit conditions. A similarly small
photopotential was observed (as had been reported earlier [21]) when the mem-
brane was illuminated with light of wavelengths 450 nm (1-4 mV) and 675 nm
(0—5 mV). The characteristics of the response at the three different wave-
lengths were the same: a fast rise in the potential, reaching a maximum value
under constant illumination and a relatively fast decay when the light' was
turned off. An open-circuit photovoltaic action spectrum was difficult to
record due to the small values of the photopotential. The difficulty was
increased by the bleaching of the chlorophyll pigments which became evident
after eight or ten subsequent illuminations.

Since the photopotentials were very small, it was decided to impose some
asymmetrical conditions across the membrane to see if the response could be
enhanced.

A membrane was formed and after it had reached the ‘black’ state (mem-
brane dark resistance approx. 1 - 108 $2), a potential of 35 mV was applied

436

across the membrane. When the membrane was illuminated with 300 nm radia-
tion, a closed-circuit photoresponse of 14 mV with opposite polarity to the
applied potential was observed. A spectrum of the closed-circuit photoresponse
in the ultraviolet as well as in the visible regions was recorded. It should be
mentioned that, under the conditions of externally applied potential, the
bleaching of the chlorophyll pigments was much stronger than that under the
open-circuit conditions. It was therefore difficult to record a complete spec-
trum for the whole region (250-700 nm) for a single membrane. The
reproducibility of the photoresponses for different membranes at a particular
wavelength was poor. On the other hand, if the photoresponses were obtained
for a single membrane at three different wavelengths only (e.g., 300, 450 and
675 nm) the relative ratio of their values had good reproducibility. Calculating
similar ratios for different wavelengths and comparing them with those
obtained for the open-circuit measurements it was found that the two spectra
were fairly similar.

Since addition of electron acceptors to one side of these membranes is
known to enhance the photoresponse in the visible region [22], it was decided
to investigate its effect under ultraviolet illumination. A bilayer lipid membrane
was formed on the Teflon cup and after reaching the black state, 250 ill of
0.1 M FeCl, solution was added to one side of the membrane (final concentra-
tion 3 - 10‘3 M). It was observed that the value of the membrane resistance
increased from 1 - 108 to approx. 5 - 108 52. When the membrane was
illuminated with light of 300 nm wavelength, an open-circuit photopotential of
up to 100 mV could be obtained. The photopotential was negative on the side
containing the FeCl3. It was observed that, when FeCl3 was present on the side
of the membrane facing the illuminating radiation, the ultraviolet response was
almost zero whilst the visible response could still be observed. In photovoltaic
action spectra measured this way the ultraviolet portion was completely
removed, and at very high FeCl, concentrations (above 1 - 10'3 M) the blue
peak at 450 nm was observed to decrease relative to the 670 nm peak. This was
found to be due to the strong absorption of FeCl, on the 200—400 nm region,
thereby causing a reduction in the intensity of the ultraviolet light incident on
the membrane. The results of control experiments involving the simultaneous
presence of FeCl3 on both sides of the membrane would then not be valid. No
photoresponse is obtained when equal concentrations of FeCl, are present at
the same time on opposite sides of the membrane under visible illumination
[22]. By extension we expect the same to hold true under ultraviolet illumina-
tion.

A photovoltaic action spectrum in the range 250—700 nm of a chloroplast
extract membrane in the presence of FeCl, is shown in Fig. 1.

The effect of electron donors was also investigated by adding 250 ill of
0.1 M ascorbate to one side of the membrane. When the membrane was
illuminated by light with a wavelength of 300 run an open-circuit photo-
response could be observed. An action spectrum was recorded (see Fig. 2) in
the range 250—700 nm. The side containing the ascorbate was positive. Similar
to the experiments involving FeCl3, it was observed that if ascorbate was
present in the outer chamber, the ultraviolet portion of the spectrum was much
smaller and had a different shape to that when ascorbate was present in the

437

 

l5.0 "

Photopoteniiol (mV)

|0.0 *'

5.0 "

 

 

 

 

zoo 360 460 500 660 700
Wavelength (nm)

 

Fig. 1. Curve I (X X): open-circuit photovoltaic action spectrum of a chloroplast-bilayer lipid mem-
brane in the presence of FeCl3. Curve II (C- i‘): absorption spectrum of chloroplast-bilayer lipid
membrane-forming solution. Curve III: Curve 1 X4.

 

inner chamber. On the other hand, the visible photoresponses were similar in
shape and magnitude when ascorbate was present in either chamber. Again this
was observed to be due to the absorption of ultraviolet light by ascorbate.

 

 

 

 

80’
is
S‘ ‘5
.6. 6°' 5
§ §
no .5
§ 8
.. 2
0
g 40* i
15
20-
zoo 300 400 500

Wavelength (nm)

Fig. 2. Comparison of the action spectrum of a chloroplast-bilayer lipid membrane in the presence of
ascorbate (X X) and in the presence of FeCl3 (0 ii).

 

 

438

Therefore, the results of control experiments where ascorbate is present simul-
taneously on both sides of the membrane are not valid for ultraviolet measure-
ments. However, these experiments have been performed for visible illumina-
tion and they show that under symmetric conditions there is no photoresponse
[4]. We expect the same to hold true for ultraviolet illumination.

Since FeCl, and ascorbate both absorb in the ultraviolet, experiments could
not be successfully performed with the two modifiers present simultaneously
on opposite sides of the membrane under ultraviolet illumination. These experi-
ments have already been performed for visible light and they show that a
maximum photoresponse is obtained under these conditions [4]. For the same
reason, all the results reported above are for FeCl, or ascorbate in the inside
chamber.

In another set of experiments, membranes made up of a mixture of phospho-
lipids and cholesterol in n-octane (see Materials and Methods) were formed in
0.1 M acetate buffer at pH 5. After the membrane had thinned to the black
state, FeCl3 or ascorbate was added to the inner chamber to a final concentra-
tion of up to 5 - 10'3 M. Larger concentrations made the membranes unstable.
The presence of FeCl, was observed to increase the dark resistance of the
membrane from approx. 1 - 108 to approx. 1 - 109 $2. Ascorbate did not show
a definite effect on the dark resistance of the membrane. When these mem-
branes were illuminated with ultraviolet or visible light in the presence of
FeCl, or ascorbate, no photoresponse could detected. As mentioned earlier, we
also observed an increase in the dark membrane resistance for the chloroplast
membrane in the presence of FeCl,. On the other hand, it has been reported
that the dark membrane resistance Of an oxidized cholesterol membrane can
decrease by up to two orders of magnitude in the presence of FeCl, [23]. No
experiment was performed with the non-pigment components of the chloro-
plast extract in the presence of FeCl, or ascorbate.

Bacteriorhodopsin-containing bilayer lipid membranes

Membranes were formed in a 75 mM KCl + 25 mM CaCl, aqueous solution
and after the membranes had thinned to the black state, lipid vesicles con-
taining purple membrane were added to one side of the bilayer lipid membrane.
After 5 min, small photoresponses could be observed when the membranes
were illuminated with visible light (560 nm). After 45—50 min, when the
photoresponses had leveled off to their maximum values, the membranes were
illuminated with ultraviolet light (300 nm) and photoresponses of approx. 25
mV were observed. An open-circuit photovoltage action spectrum was obtained
in the range 250-700 run. When the membranes were illuminated with ultra-
violet light in the absence of the vesicles no photoresponse could be observed.
A decrease in the membrane dark resistance of up to one order of magnitude
was observed for the membranes in the presence of the vesicles.

Bilayer lipid membrane with chymotrypsin in the aqueous solution

Membranes were formed in 1 - 10" M KCl. After the membrane had thinned
to the black state, 200 pl of a 5% chymotrypsin solution were added to one
side of the membrane (vol. 8 ml). After 10 min a very small photopotential was
Observed when the membrane was illuminated with ultraviolet light. After 40

439

min a maximum photoresponse of approx. 10 mV was observed. An open-
circuit photovoltage action spectrum was recorded in the range 250—450 nm.
No photoresponse was Observed when the membrane was illuminated with
visible light. A bilayer lipid membrane was formed in a 10 mM KCl solution
and a smaller photoresponse than before was observed when the same amount
of chymotrypsin was added. The photoresponse was still smaller in a 0.1 M KCl
solution. No photoresponse could be observed in the absence of chymotrypsin.

Discussion

Bilayer lipid membranes formed from chloroplast extracts exhibit photo-
responses when illuminated with ultraviolet light. Small open-circuit photo-
responses were observed under chemically symmetric conditions presumably
due to differences in the absorption of the incident light between the two
layers of the membrane. The fact that these photoresponses are not due to
thermal effects is supported by the fact that the light intensity incident on the
membrane was very small and also by the fact that the ‘thermoelectric power’
of the chloroplast membrane was also very small, approx. 60 uV/K over the
range 15—40°C [4]. Closed-circuit photoresponses were also obtained under an
externally applied electric field. This photoresponse could be interpreted as
being due to a photoconductive effect, however, since an open-circuit photo-
response had also been observed, the photoresponse cannot be totally due to a
change in the conductance. Furthermore, the fact that both action spectra were
very similar indicates that the photoconductivity effect should be very small.

The action spectra obtained under closed- and open-circuit conditions also
followed the absorption spectrum of the chloroplast membrane in the range
250—700 nm. This indicates that the primary event involved in the production
of the potential is the absorption of light by the pigments present in the mem-
brane.

In the presence of FeCl, and ascorbate, the action spectra are very similar
(see Fig. 2) and both showed an enhancement of the ultraviolet part relative to
the visible part (see Fig. 1). The fact that the photoresponse in the ultraviolet
can be enhanced by the addition of electron donors or acceptors suggests the
possibility that the nature of the charge separation in the ultraviolet involves
energy transfer and redox reactions on opposite sides of the membrane.
Coupled redox reactions taking place on opposite sides of the membrane sug-
gest the possibility of electronic conduction through the membrane. That
electronic conduction takes place in these bilayer lipid membranes was defini-
tively established [4,11] after experiments were performed in which the mem-
branes were excited by light flashes of microsecond duration. Photosensitiza-
tion experiments involving this particular chloroplast bilayer lipid membrane
were also successively performed using different dyes [4,18]. In these experi-
ments, both types of sensitization, intrinsic as well as extrinsic were observed.

In the case of the bilayer lipid membrane containing purple membrane, the
fact that the action spectrum in the ultraviolet follows the absorption spectrum
of the bacteriorhodopsin in that region (see Fig. 3) indicates that those species
in the bacteriorhodopsin molecule which absorb ultraviolet light are involved in
charge separation. Three possibilities are: ( 1) charge separation takes place by

440

 

 

 

 

I . 0.7
. 0.6
50 . . 0.5
J 0
4O? ’r‘d‘ 0.4. g
3 r
g as" I, g
E 30 n ‘1 ‘I‘ I], as
s
§ 20)- “ t‘ 02
g \P--4-‘*""“r’
IO r i OI
\
200 300 400 500 600 700
Wavelength (nm)
Fig. 3. Curve! (O—O): absorption spectrum of purple membrane from Halobacterium halobium.
Curve II (X ------ X): open-circuit photovoltaic action spectrum of a bilayer lipid membrane-containing

purple membrane.

direct action of the absorbing species; ( 2) charge separation is mediated by the
absorbing species probably through energy transfer [26] or (3) both 1 and 2
are occurring simultaneously (some recent results give strong evidence for the
second or third possibility). From Fig. 3 it can be seen that the relative
intensity of the ultraviolet part of the spectrum is about half that of the visible.
Therefore, not all the ultraviolet absorbing species in the protein can induce
charge separation. This indicates that the nature, relative position, orientation
and environment of the absorbing species may be involved. '

The separation of charges across the membrane due to direct absorption of
ultraviolet light by a protein was also observed in the case of the association of
chymotrypsin with the bilayer lipid membrane. Again, in this case, the photo-
electric action spectrum followed the absorption spectrum Of the chymotrypsin
(Fig. 4). The fact that the protein binds to the membrane is suggested by the
fact that the photoresponse is not observed immediately after addition of the
protein but after some time and that it increased with incubation time. This is
also suggested by the fact that no photoresponse is observed at high salt con-
centrations (above 0.1 M KCl) which can be explained in terms of shielding of
the charges present in the membrane and protein interface by counter-ions.

There are a number of questions that need to be addressed: (i) the number
of photoactive species in the membranes; (ii) the orientation, location and con-
formation of constituent molecules in the bilayer lipid membrane, and (iii) the
mechanisms by which energy and charges are transported. The answers for the
first two questions may be found in a number of papers [4,6,8,22]. In the case
of chlorophyll, for example, the maximum concentration is approx. 1 - 10"°
mol/cm’, with the porphyrin plate at an angle of approx. 45° with respect to
the plane of the membrane. The hydrophobic phytol chain is anchored in the

441

 

 

 

 

 

5.0 l
4.0 " i 0.8
S‘
E 3.0 ~ . 0.6
§ 0
.8. i
O
8 s
E 2
°' ao . 0.4
I.o - . 0 2
II
200 250 300 350
Wavelength (nm)
Fig. 4. Curve I (O-—O): absorption spectrum of chymotrypsin in distilled water. Curve II (X ------ X):

open-circuit photovoltaic action spectrum of bilayer lipid membrane to one side of which chymotrypsin
was added.

lipid bilayer. The porphyrin ring of the chlorophyll molecule is depicted to be
among the polar head groups of lipid molecules in contact with the aqueous
solution [24].

In the bacteriorhodopsin and chymotrypsin bilayer lipid membrane systems
described here, no definite information on molecular organization as yet is
available.

As mentioned earlier, in the chlorophyll-containing bilayer lipid membrane,
the charge carriers are electrons [4,11] and the observed photoelectric phenom-
ena are primarily due to the membrane/electrolyte biface [4,5,27]. The
observed photovoltage is influenced by the redox compounds (electron accep-
tors and donors) present in the bathing solution [24,28]. To explain the
observed photoelectric effects, the pigmented bilayer lipid membrane has been
considered to be an organic semiconductor as shown in Fig. 5. The situation is
quite analogous to that of a Schottky barrier except there are two interfaces
[24]. That the observed photoelectric phenomena are primarily due to the
bilayer lipid membrane/electrolyte biface and not the plateau-Gibbs border has
been definitely established [4,5,27]. In Fig. 5, one side of the membrane is
depicted as p-type, hence it acts as a photocathode and the other side is n-type
(photoanode). In the bilayer lipid membrane/electrolyte interface, the aqueous
solutions play the role of the metal. Thus, the energetics of the interface at

442

MEMBRANE

SOLUTION SOLUTION

flax-..
‘9. 3y
E,.,,,——-——--L-f

 

 

E». 'r— £9 1 ‘E 5.070

H"

l

I

I
11.
‘1'

 

 

Fig. 5. Schematic diagram of a pigmented bilayer lipid membrane separating two aqueous solutions. (A)
the membrane is in equilibrium with redox compounds in the bathing solutions in the dark. (8) the
bilayer lipid membrane is under illumination.

equilibrium in the dark and in the light under open-circuit conditions are
shown in Fig. 5A and Fig. 5B, respectively. E; is the band gap, E, is the Fermi
level. E A. ,A (or ED. m) is the electrochemical potential of the solution, Ecs and
Eva are the respective conduction and valence band positions. The symbols for
the right-hand side are primed. The observed photovoltage, AV(EV + E(,) is
obviously influenced by the redox compounds (electron acceptors and donors)
present in the bathing solutions [24,28]. It should be pointed out that photo-
generated electron-hole pairs in the bilayer lipid membrane result in current
flow, except, unlike the metal in a Schottky-type cell, the aqueous solution is
not an electronic conductor. Therefore, redox reactions must take place at the
two interfaces in order to complete the circuit. A more detailed discussion on
redox reactions and on the semiconductor model of charge separation and
transport in pigmented lipid membranes has been published elsewhere [28,29].
It should be mentioned that in the case of bacteriorhodopsin-containing bilayer
lipid membrane, photogenerated H+ (and OH'), owing to their small size, are
unique and may act as primary charge-carriers in the membrane.

The major finding reported in this paper is that conventional ultraviolet as
well as visible spectroscopy can be readily combined with photoelectric mea-
surements in the bilayer lipid membrane system. The resulting technique, called
bilayer lipid membrane ‘photoelectrospectrometry’ [30], should offer an
approach to the study of energy transfer between lipids and proteins [31] and

443

membrane reconstitution experiments. Some experiments of this type are in
progress.

Conclusions

Charge separation can occur in pigments and proteins present in bilayer lipid
membranes when they are illuminated with ultraviolet light. This allows for
ultraviolet photoelectric action spectra to be recorded. The photoresponses
obtained in the ultraviolet and visible regions for a chloroplast extract bilayer
lipid membrane and bilayer lipid membrane-containing bacteriorhodopsin
appear to be due to independent processes and exhibit different quantum effi-
ciencies. Chymotrypsin can bind electrostatically to a bilayer lipid membrane
and this association allows for charge separation across the membrane when it
is illuminated with ultraviolet light. A photopotential action spectrum of the
protein and bilayer lipid membrane follows the absorption spectrum of the
protein. This result suggests the possibility of studying lipid-protein inter-
actions in membranes through the extension of the spectroscopic studies of a
bilayer lipid membrane into the ultraviolet region which is where most proteins
absorb light.

The fact that the ultraviolet part of the spectrum, in the case of the chloro-
plast extract bilayer lipid membrane, is enhanced by the asymmetric addition
of electron acceptors and donors suggests the possibility of redox reactions
taking place on opposite sides of the membrane. The technique, termed photo-
electrospectrometry, is at least four orders of magnitude more sensitive than
absorption spectroscopy and is useful in the investigation of lipid-protein inter-
actions and energy transfer studies in membranes.

Acknowledgements

This work was supported by a National Institute of Health Grant
(GM-14971). We thank Dr. J. Higgins for his invaluable help and stimulating
discussions.

References

1 Schreckenbach, 'I‘. (1979) in Photosynthesis in Relation to Model Systems (Barber, J., ed.), chapter 6,
Elsevier/North-Holland Biomedical Press, New York
2 Bangham, A.D. (1975) in Cell Membranes (Weissmann, G. and Claiborne, R., eds.). pp. 24—34, HP
Publishing Co. Inc., New York
3 Andersen, 0.8. (1978) in Membrane Transport in Biology (Tosteson, D.C.. ed.), chapter 11, Springer-
Verlag, Berlin
4 Tien, H.T. (1974) Bilayer Lipid Membranes: Theory and Practice. Marcel Dekker, New York
5 Hang, F.T. (1977) J. Colloid Interface Sci. 58, 471—496
6 Steinemann, A., Alamuti, N., Brodmann, W., Marschall, O. and Laeuger, P. (1971) J. Membrane Biol.
4, 284—294
7 Mangel, M., Berna, 0.8. and Ilani, A. (1975) J. Membrane Biol. 20, 171-1 80
8 Steinemann, A., Stark, G. and Laeuger, P. (1972) J. Membrane Biol. 9, 177—194
9 Master. 8.8. and Mauzerall, D. (1978) J. Membrane Biol. 41, 377—388
10 Hess, M. (1977) Naturwissenschaften 64, 94
11 Kobamoto, N. and Tien, H.T. (1971) Biochim. Biophys. Acta 241, 129—146
12 Schadt. M. (1973) Biochim. Biophys. Acta 323. 351—366
13 Fesenko. E.E. and Lyubarskiy, A.L. (1977) Nature 268, 562—563

444

25
26

W

28

29
3O
31

Shieh, P.K. and Packer. L. (1976) Biochem. Biophys. Res. Commun. 71, 603—609

Karvaly, B. and Dancshazy, Z. (1977) FEBS Lett. 76, 45—49

Herrmann, T.R. and Rayfield, G.W. (1978) Biophys. J. 21, 111-123

Pohl, R.H. and Teissie, J. (1975) Z. Naturforsch. 300, 147—1 51

Huebner, J.S. (1978) J. Membrane Biol. 39. 97—132

Ullrich, H.M. and Kuhn, H. (1972) Biochim. Biotha. Acta 266, 584—596

Blok, M.C., Hellingwerf, K.J. and van Dam, K. (1977) FEBS Lett. 76, 45—50

Tien, H.T. (1968) Nature 219, 272—274

Loxsom, F.M. and Tien, H.T. (1972) Chem. Phys. Lipids 8, 221—229

Karvaly, B. and Pant, H.G. (1972) Stud. Biophys. 33. 51—58

Tien, H.T. (1979) in Photosynthesis in Relation to Model Systems (Barber, J., ed.). pp. 116—173.
Elsevier/North-Holland Biomedical Press, New York

Brown, J.S. (1977) Photochem. Photobiol. 26. 319-336

Seely, G.R. (1977) in Primary Processes of Photosynthesis (Barber, J;, ed.). pp. 1—53. Elsevier/North-
Holland Biomedical Press, New York

Dancshazy, Z., Ormos, P., Drachev, L.A. and Skulachev, V.P. (1978) Biophys. J. 24, 423—428

Tien, H.T. (1978) in Photosynthetic Oxygen Evolution (Metzner, H., ed.). DD. 411—438, Academic
Press, New York

Tien. H.T. (1980) Separation Science and Technology. M. Dekker, Inc., New York

Van, N.T. and Tien. H.T. (1970) J. Phys. Chem. 74, 3559—3568

Schreckenback, T., Walckhoff, B. and Oesterhelt, D. (1978) Biochem. 17, 5353—5359

APPENDIX B

 

APPENDIX B

Bioelectrochemistry and Bioenergetics .6 (1979) 509—524
f. Electroanal. Chem. 104 (1979) 509—524
. © Elsevier Sequoia S.A., Lausanne - Printed in Italy

284- - H. Halobium: I. In vitro Studies

by JOHN HIGGINS, JOSE R. LOPEZ and H. TI TIEN

Department of Biophysics, Michigan State University, East Lansing, Michigan
48824 (U.S.A.)

Revised manucript received July 20th 1979

Summary

I. Purple membrane fragments (PM) from H. halobium were in-
corporated into lipid membranes either directly or via liposomes. A
photoresponse was detected when liposomes containing PM were fused
with lipid membranes. Low—resistance membranes or membranes shunted
with an external resistor of 109 ohms showed decay of the initial light
response to some equilibrium value in the light in both presence and
absence of octadecylamine. The light response could be abolished by
the addition of a sufficient amount of triethylamine to either side of the
membrane.

2. The photovoltage action spectra of bilayer lipid membranes con-
taining PM either directly or via liposomes were measured, and found to
follow the absorption spectrum of bacteriorhodopsin.

3. Liposomes containing bacteriorhodopsin (BR) in PM extract of
H. halobium were fused to one side of planar lipid membranes. The
photOpotential resulting from flash excitation rose and fell as the sum
of three exponentials with time constants for the leading edge of 30 i IO
us and 35 i IO ms. The decay time constant of the photopotential was
840 ms, a value consistent with the membrane time constant given by
the membrane dark resistance and capacitance.

 

Introduction

H. halobium, in which the so—called purple membrane is located, was
first noted on dried salted fish in about 1900 in Scandinavia. This bacte-
rium required a near—saturated salt solution for growth. The biochemical
aspects of extreme halophilism displayed by these organisms have been
reviewd by LARSEN.1 The purple color of H. halobium is due to a rho-
dopsin—like complex in the plasma membrane, termed bacteriorhodopsin,
that develops when the cells are grown in the absence of a suitable source
of nutrients or at low oxygen tension.2 Bacteriorhodopsin, upon absorp-

157

510 Higgins,‘Lopez and Ti Tien

tion Of light, undergoes a photoreaction cycle which involves the translo-
cation of protons, thereby generating a proton gradient that can be used
as the driving force for ATP synthesis in accordance with the chemios-
motic hypothesis.3 Recent work on H. halobium and its derivatives, purple
membrane (PM) and bacteriorhodopsin (BR), have been comprehensively
reviewed by OEerRHELr‘ and by SCHRECKENBACH.5 Studies by spec-
troscopic techniques have elucidated the photochemistry Of BR and
PM!”8 Other studies include phase transitions by differential scanning
calorimetry,“ light—induced pH changes and ATP synthesis,2.1°-“ recon-
stitution experimental-““17 and incorporation Of BR and PM into artificial
lipid bilayers of both planar (bilayer lipid membranes or BLM) and
spherical (lipid microvesicles or liposomes) configuration as well as milli-
pore filtersm'“

Although much is known about the purple membrane and its de-
rivative, bacteriorhodopsin Of H. halobium, the mechanism Of proton
translocation by BR in the light remains Obscure. However, it is generally
accepted that the purple membrane (bacteriorhodopsin) acts as a light—
driven proton pump and transports protons vectorially from inside the
cell to the surrounding medium. That is, during the course of each pho-
toreaction sequence, BR first releases protons on the extracellular side Of
the purple membrane and then picks them up on the cytoplasmic side.”
This generally accepted direction Of proton movement, however, has been
recently questioned on the basis Of published findings.25 We have, there-
fore, carried out both in vivo and in vitro studies on H. halobium, regarding
both the direction Of proton movement and certain physical properties
of BLM and liposomes containing the purple membrane hitherto unre-
ported.“

Experimental

1)! aterials

Halobacterium halobium was grown in two—liter flasks under intense
illumination from fluorescent tubes using essentially the method described
by BECHER and CASSIM.” The medium contained in I000 cm“: NaCl,
250.0 g; magnesium sulfate, 20.0 g; KCl, 2.0 g; CaCl._, - 2 H2O, 0.2 g;
sodium citrate, 3.0 g ; SIGMA peptone No. P—8388, I0.0 g or DIFCO Yeast
Extract (OI27—03), 5.0 g. Each culture was inoculated with 200 cm3
of cells from a previous culture. The cells were grown with 15 cubic feet
Of air per hour bubbled through the culture for 5 days or until the end
of log phase. Then the cells were aerated at the rate of 3 cubic feet per
hour for 3 days. After that the cells were harvested. They were spun
down and washed twice in basal salts (first 4 items listed above for the
medium) before use.

The purple membrane (PM) was isolated by dialyzing Halobacterium
halobium cells overnight against 0.1 M NaCl and then centrifuging the

H. halobium : I. In vitro Studies 511

material from the dialysis bag at 31,000 g for 30 to 45 minutes. The red
supernatant was then decanted and the pellets used as purple membrane.

The membrane—forming solution was prepared in the following
manner: lipids were extracted from whole cells with 2: I chloroform:
metanol. The cells were homogenized with the chloroform : methanol in
a blender for 3 minutes. The preparation was then filtered (NO. 42 paper)
and the filtrate flash—evaporated and redissolved in n-octane or n—decane.
These lipids were combined in a I: I mixture with a 2 % solution of
phosphatidyl choline in n—octane or n—decane to make the BLM-forming
solution.

Incorporation of purple membrane (PM)

For planar bilayer lipid membrane (BLM) studies, either the PM
was incorporated directly into the membrane—forming solution or lipid
microvesicles (liposomes) containing PM were allowed to fuse with the
BLM. The liposomes were made by drying enough egg lecithin (SIGMA)
to form a I % solution on the inside of a round—bottomed flask. Purple
membrane dissolved in 0.15114 KCl was then added and the solution
dispersed on a vortex mixer in the presence of small glass beads. This
suspension was then sonicated (BRANSON model W—I4OD) at high power
(60—70 watts) under nitrogen atmosphere and at low temperature (ice
bath) for I hour (15 seconds of sonication followed by 45 seconds coooling
interval).

Standard bathing solution and Optical cells28 held the membranes
for the other measurements. A planar membrane was formed and elec-
trical and photoelectric a1 properties were measured in the usual manner.”
Only for experiments in which the PM was incorporated into the mem-
brane were light—induced voltages Observed. Action spectra were Obtained
using a halogen lamp as a source of excitation. The white light from the
lamp was focused by means Of a lens onto a BAUSCH and LOMB high
intensity grating monochromator. The monochromatic light was then
focused upon the BLM in the teflon cup by means Of a microscope lens.
The photovoltage was measured with a KEITHLEY INSTRUMENTS Model
6IO BR electrometer and the spectra were recorded Of a KEITHLEY IN-
STRUMENTS Model 370 recorder. Flash illumination of 10° lumens/cm2
from a xenon flash lamp and associated circuitry (GENERAL RADIO
STROBOSLAVE, Type 153g—A and lamp) was used to study fast photo-
responses. Electrical measurements were made through calomel elec-
trodes connected by short leads to a high impedance (1012 Ohms), unity
gain buffer amplifier Of time contant 8 us. Membrane potential was
Observed and recorded on a storage oscilloscope (TEKTRONIX R503I)
and on a chart recorder (BAUSCH and LOMB VOM6). For all electrical
measurements, the electrode immersed in the inner cup went to the high
side of the buffer amplifier or electrometer, so that the sign of the voltages
is that Of the inner chamber.

When PM was incorporated by way Of liposomes the usual proce—
dure was to form the membrane, add the liposomes to the inner chamber,

512 Higgins, Lopez and Ti Tien

and wait until a stable photovoltage was reached. This waiting time
depended upon the concentration of liposomes and stirring Of the bath;
it varied from IO minutes at the highest concentration when the bath
was stirred to a maximum Of 250 minutes when the concentration was
diluted IO—fold and the bath was not stirred.

Photoeflects in Purple Membrane - BLM

The Observed photoeffect when liposomes containing PM are fused
with BLM’S can be understood in terms of protons being pumped from
the inner chamber (to which liposomes were added) to the outer chamber
in the presence of light. Fig. I Shows the Open—circuit photoresponse for
a membrane containing 0.0025 % octadecylamine (ODA). Previously,

 

 

l 2 3 4 5
O l I I I I
Time (min)
on
-10 '- \ Jr
3:
E
-20~ I
.3
E"
2
-30 - g
'40?

 

off

Fig. I.

Photoresponse seen for BLM made from I : 1.2 91, phOSphatidyl choline and lipid extract
from H. halobium with 0.0025 93 octadecylamine in n—octane. Rm was 7 x 10’ Q. Bathing
solution was 75 mil/I KCl and 50 mM CaCl,, and 100 mm3 (51.1) liposomes with PM added to
the inner chamber were allowed to fuse for 4 hours. The dark potential was —Io mV
and the photoresponse was —30 mV. NO light—decay or dark-overshoot were seen.

KARVALY and DANCSHAZY ‘3 have reported that the presence Of ODA
in the bilayer lipid membrane resulted in a drastic increase of the protein
binding to the BLM. Further, they reported that BLM containing the
PM exhibited both a large—amplitude photovoltaic effect and photocon-
duction. The polarity Of the compartment with the PM was negative
upon illumination for all the systems investigated, indicating that at
least for the PM—BLMs protons probably move across the membrane as
suggested by several authors (for reviews, see Refs. 4 and 5). The am-

 

H. halobium': I. In vitro Studies 513

plitude 0f the photovoltaic response was Of the order Of 20 to 60 mV,
depending upon the concentration of the incorporated PM and the in-
tensity Of the light. Fig. 28 shows that when the membrane Of Fig. I
is shunted by a 10’ Q resistor both light—decay (decay of voltage under
illumination) and dark-overshoot potential (upon removal Of illumina-
tion) can be seen. Also, the lower the resistance the lower was the maxi-
mum response. We have fully confirmed the Observations Of KARVALY
and DANCSHAZY 13:22. Further, Fig. 2 (A and C) shows the effect of an
applied potential. Making the inner chamber, containing the PM lipo-
somes, negative with respect t0 the outer chamber decreased the maximal
response while enhancing the light—decay and dark—overshoot. Positive
applied potentials enhanced the maximal response while reducing the
light—decay and the dark—overshoot potentials.

 

 

 

>
E
Q
60 5
off
i
d. T 4F
on
A 3 C

Fig. 2.

Photovoltages of shunted BLM. Same BLM and bath as described in Fig. I shunted by
10’ Q external resistor and subjected to applied potentials. The up arrows mean the light
is turned on and the down arrows indicate the light is turned Off. Three waveforms show
the effects of (A) +54 mV (8) 0 mV, and (C) -—54 mV applied potential on the photo-
response. All the photovoltages are negative in sign.

In the absence Of octadecylamine both the Open—circuit and shunted
BLM gave photovoltage waveforms that also exhibited light—decay and
dark-overshoot potential, with shapes like that of Fig. 28 (see also Fig. 5).
Both effects were enhanced by lowering in either membrane resistance
(Rm) or shunt resistant (R,).

The effects of octadecylamine, resistance, and applied potential
can be understood if the light—decay and dark—overshoot are due tO the

514 Higgins, Lopez and Ti Tien

variable permeability Of the membrane to protons and the possible presence
Of an external circuit for electrons. The electrons, attracted by the elec-
trostatic charge of the increased proton concentration in one chamber,
will flow via the external circuit (shunt resistor) into that chamber. In
the absence of an electronic pathway (no external Shunt), the ODA
prevents light-decay by providing a positively charged membrane which
prevents protons from flowing back across the membrane. The high
resistance membranes tend to do the same, and the positive applied po-
tentials also hinder the back proton flow. Triethylamine, an electron
and proton donor, was found to short out the photoresponse when added
to either the inner or the outer chambers. It is thought, according to
the above scheme, that triethylamine could be transporting or releasing
either protons or electrons to short out the response.

Spectroscopy of purple membrane—8L1”

Planar bilayer lipid membranes containing purple membrane (PM)
incorporated by liposome fusion on one side Of the membrane exhibited
a large amplitude fast photovoltage response between 15 and 50 mV at
constant monochromatic illumination, depending on the amount Of li-

 

 

«400
I50-
~3eo
A 4300
ii:
IOOh E «250
g % <zoo
«ISO
50- 0
/°/ boo
/
D/
// .50
1°4/l/ I A (nmll \\
300 400 500 600 7%

 

Fig. 3.

Open—circuit photovoltaic action spectra. PM has been incorporated into BLMS by two
different methods: vesicle fusion (0, right—hand scale) and dissolving directly into the
membrane forming solution (0, left—hand scale). All other experimental conditions were
the same for both membranes: basic BLM forming solution of I : r 2 0,1, PC in decane:
H. halobium extract, aqueous bath 75 mM KCl and 50 mM C210,, and Rm range of 5—8 x
x IO“ Q.

H. halobium : I. In vitro Studies 515

posomes added and on the wavelength used. The side containing the
liposomes with PM was always negative upon illumination. The pho-
tovoltage action spectrum from this PM—BLM system is shown in Fig. 3.
BLM containing PM which was incorporated directly into the membrane
forming solution also exhibited a fast photovoltage effect upon illumina-
tion with monochromatic light. The photovoltage response was between
5 and IO mV. The reason for a smaller photoresponse is probably due
to the fact that the amount Of PM incorporated by liposome fusion was
larger than in the case Of incorporating PM into the membrane forming
solution, given that the PM is slightly soluble in organic solvent.

1

25.0

A

20.0

Photoemf (mV)

1 ISO

 

 

 

,’ A (nm) e
300 450 500 600 700

 

Fig. 4.

Comparison of spectra. One action Spectrum of Fig. 3 for PM vesicle fusion (0, right-
hand scale) is compared with the absorption spectrum of bacteriorhodopsin in aqueous
dispersion (solid line, left—hand scale).

The photovoltage action spectra Of both types Of PM—BLM were
similar (Fig. 3), and resemble very much the absorption spectrum of
bacteriorhodopsin in aqueous solution except for the region between 350
nm and 450 nm (Fig. 4). This has been explained by DANCSHAZY and
KARVALY18 as being due to shielding or quenching by carotenoids, which
absorb in this region. Recent spectroscopic investigations on purple
membrane fragments layered on a glass surface possess definite shoulders
or even maxima at the wavelengths indicated above.29 This means that
a certain amount of the absorbed light may be dissipated or emitted as
fluorescence.

516 Higgins, Lopez and Ti Tien

The time course of the PM-BLM photoresponse as a function of
wavelengths was monitored. Fig. 5 (A and B) illustrate the light—induced
voltage at 560 nm and at 411 nm, respectively. The complicated wave-
form shown in Fig. 5.4 will be discussed in conjunction with the flash
experiments described in the next section.

A 560nm _ 8 411nm

 

 

Fig. 5.

Open-circuit photopotential Of PM-BLM. The membrane was illuminated at two wave—
lengths and equal intensity (about 5 mW/cm'). Dark membrane potential is zero. PM
incorporated by fusion of liposomes with PM to BLM. R,” was 5 x 10' Q. and membrane
and bath solutions as given in Fig. 3. Light on and Off is indicated by arrows. Ordinate
is the potential of the inner chamber.

All the above data taken together, it can be concluded that the
resemblance between the photovoltage (action) spectrum and the absorp-
tion spectrum establishes that the Observed photoeffect is due to the pres-
ence Of PM in the BLM.

The successful incorporation Of PM into BLM and the preliminary
photoelectronspectroscopic investigations on them give impetus to in-
corporate charged proteins into lipid bilayers and may make even broader
the application Of the BLM technique in studying the function Of proteins
in membrane processes as well. The data on the wavelength dependence
of photoresponses clearly demonstrate again the unique power of photo-
electric spectroscopy in studying photobiological phenomena at the mo-
lecular level.”29

Flash excitation of purple membrane-BUM

Typical photovoltages of PM—BLM to a 3 us xenon flash are shown
in Fig. 6. Purple membrane was incorporated by fusion Of liposomes and
BLM. The Open—circuit photovoltage, designated as V,,(t), appears to
consist of two voltages. One component (V4) is proportional to the po-
larization of the membrane by the net dipole electric field of the double
layer produced by the finite lifetime of protonated and unprotonated
forms Of bacteriorhodopsin (BR) as it pumps protons. The other com-

H. halobium .' I. In vitro Studies 517

 

 

 

 

l N 0.1 mS/div
_ I ms/div
*1; /0.1 s/div
. —- ‘10 ms/dlv
. l S/div

l L 1 4L 1 1 1 l l l t l l l

 

Fig. 6.

Photoflash potential. Tracings show the oscilloscope displays of the Open-circuit photo-
voltage Of PM-BLM following intense 3 us xenon flash (indicated by arrows). One minute
interval between flashes. PM incorporated by fusion of liposomes with PM to BLM. SO-
lutions as in Fig. 3. Rm was 2 x 10' Q. Ordinate is the potential of the inner chamber.

ponent (V,) arises directly from the proton current, Of which part estab-
lishes an Ohmic potential drop across the membrane resistance (R...)
while the remainder charges the membrane capacitance (Cm). In similar
and related pigmented bilayer lipid membrane systems it has been
proposed that the Observed photopotentials are best explained by treating
the BLM together with its two co—existing solution (membrane interfaces
as a parallel RC circuit and a generator, which is produced by light, con—
nected across the membrane.23v3°'3° In the present case, owing to the
fusion Of liposomes to the membrane, the system is better represented by
an equivalent circuit shown in Fig. 7A. The battery and resistor enclosed
by dotted lines represent the PM proton pump, which acts as a non—
ideal potential—generator.“ The validity Of a similar model circuit for
the system Of PM~liposomes fused tO planar BLM has been demonstrated,“
the only difference being the representation Of the generator by a voltage—
dependent ideal current generator. But this is electrically equivalent to
the non—ideal potential-generator (one that has an internal resistance)
shown in Fig. 7D.

The major portion Of Observed open-circuit photoresponse, V,(t),
cannot be explained by the charging and discharging Of membrane capa-
citances in Fig. 7 as the PM proton pumps turn on. First, the probable
values Of the circuit parameters, allowing for estimated density Of lipo-
somes on the planar membrane, give small dark—overshoot, defined as
the ratio Of Vd—V; to V.- (see Fig. 5A). Second, the photovoltage rise
time is limited by time constant 1,. = R.,. C". Typical values of r... are
of the order of 1 s whereas the Observed rise time of V,(t) was Of order
Of 20 ms.

 

518 Higgins, Lopez and Ti Tien

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 7.

The equivalent electrical circuit for PM liposomes fused with BLM. (A) Linear circuit
elements model the system. The generator (within dashed lines) simulating the pumping
ability of PM is variously represented by circuit elements B, C and D in the literature.
These are (B) ideal current generator with constant current Io, (C) non—ideal current gen-
erator with finite shunt resistance R0, short circuit current Io, and Open—circuit voltage
Vo = Io R0, and (D) non—ideal potential generator with series resistance R0, Open-circuit
voltage V0, and short—circuit current I0 =—- Vo/Ro. Generators (C) and (D) are equivalent.

We therefore look for an additional mechanism that could generate
biphasic Vp(t) Of comparable voltage amplitudes and time constants, a
mechanism not included in the equivalent circuit of Fig. 7. One way the
PM differs from the electrical elements Of current or potential generator
is by the generation of an electrostatic dipole field as the bacteriorhodopsin
molecule releases a proton. At the bath pH of about 7, there is evidence18
bacteriorhodopsin has an overall negative charge in both protonated and
unprotonated forms. Mutual electrostatic attraction may permit a short
lived electric dipole before the proton is drawn away. In the remaining

H. halobium : 'I. In vitro Studies 519

discussion in this section the rather complicated charge state after proton
release will be denoted by BR“ ; this is not tO be taken as necessarily a
single negative charge. Further, in the sequence Of conformational chan-
ges preparatory to proton release, BR may attain a temporary dipole
moment. Photovoltages of BLM with cyanine dyes have been attributed
to redox generated dipoles“ or to a photoinduced charge displacement.”
We extend this concept to our system, and suggest that on illumination
PM incorporated into BLM generates a dipole layer in addition to a pro-
ton current.

The proposed mechanism is as follows: Deprotonation of a BR
molecule creates a dipole, which is subsequently annihilated by proton
transport and diffusion away from BR- and by proton uptake. A dipole
layer, or double charge layer, is thus inherent in directional proton pump-
ing by all the BR in a planar membrane. The strength of the dipole layer
should depend on the proton release and uptake rates (i.e., dipole lifetime),
rates Of diffusion, and applied fields (one polarity increasing V4 and the
Opposite polarity decreasing V4). Generation Of a dipole layer establishes
immediately an electrostatic field which polarizes the medium quickly
(about I us), with a consequent rise V4 across C", at the same rate. This
rate is limited only by the rise time of the dipole layer, equivalent to the
rise time Of deprotonated BR, and would not be limited by the mem-
brane time constant. The relative amplitude Of the potential developed
by the dipole layer to that by the proton current would increase with
the dipole strength and decrease with the current associated with the
creation and annihilation of one dipole.

How can the time course of Vp(t) (Fig. 6) be interpreted in light Of
the above hypothesis? After the flash has extinguished, a number n“
Of excited bacteriorhodopsin molecules BR* remains. Each deprotonates
tO form a dipole and generate current. Later it takes up a proton to
annihilate the dipole and either

(a) create more current, if the proton uptake is on the side Opposite
release, or
(b) annihilate current if the uptake is on the same side.
Assuming (a), the reaction is

[20 hi ’3:
I + BR —>- BR* —> BR‘ —>- BR, (1)
T .

H" H"

where I is light intensity. The flash photopotential 11",,(t) is the sum Of
three potentials: Vd(t) due to the polarization charge Qp(t) on the mem-
brane in response to the dipole electric field, V1(t) = Q,(t)/C,,, due to
the charging of membrane capacitance C., by the proton current gene-
rated by proton release in equation 1 and V2(t) = 02(t)/C,,,, where 02
is the charge on C", by that part of the proton current generated by pro-
ton uptake in equation 1 :

 

520 Higgins, Lopez and Ti Tien

V,(t) = VA!) +

     

C... 1(t) + 02(5)} (2)

The time dependence of the separate terms is given by

 

 

 

 

 

     

 

 

 

     

Vat) °< 0N) 0< [6XP(— T: )—eXp(— t 1,). (3)
We . >r—expl— <.: —:.. l
V2: (t>o< exp(— 1,. ) )x—exp{— t;( —: )} (5)

 

where time constant and rate constant are related by 1/1-5 = k;. These
equations arise from the following physical considerations: C.,. may be
charged by protonic charge (Q1 and Q2) and by polarization charge (Qp).

Discharge of protonic charge can occur only by ionic leakage cur-
rent through R.,, with time constant 1... when there is no shunt resistance,
whereas polarization charge can be discharged either by leakage current,
or by collapse Of the polarizing field, with time constants 1:... and 1:2,
respectively.

The kinetics of reaction equation (1) show that theoretically an
electrical double layer is present. The question is its importance; that
is, is V4 of equation (3) comparable to or greater than V1 and V2 of equa-
tions (4) and (5)? All three components are rising and falling exponen—
tials with fast rise times initially. V.) has 11, V1 and V2 have essentially
11 and 12 because 1... is so large. Although the equivalent circuit of
Fig. 7A correctly describes the current—generating properties Of PM
(and any potentials developed by the current), it does not include any
possible dipole character Of the PM (an electrostatic phenomenon).
Consequently, this circuit may be used to examine V1 and V2, but not
V4. Clearly it shows that the rate Of rise of V1 and V2 across Cm and R".
is limited by the time constant Tm = RmCm (1 s or perhaps 100 ms), re-
gardless Of the values Of 1'1 and 1'2. Therefore, Vp(t) of Fig. 6 must include
V¢(t), the dipole potential.

Because the reaction time constants r, and r2 are much smaller
than the membrane time constant rm, they shape the leading edge of flash
Vp(t) as illustrated in Fig. 8. Proton release and uptake occur within the
sequence Of conformational changes of excited bacteriorhodopsin, BR*,
K590, L550, M412» N520, IMO, and BR570, with proton release shortly after
the formation Of M412 and proton uptake connected with the formation
of 0640.7,838,”

BR* —> M,”2 + H4", 1', = 30 us
H+ + M,“2 —> BR, 12 = 10 ms (6)

H. halobium‘: I. In vitro Studies 521

 

 

T
2
TI VA!)
.9 Mn
Tm
Tg V2“)

 

 

14, m

 

Fig. 8.

Sketch of theoretical photovoltage following flash illumination. Waveforms were deter-
mined by Equations. 3-5. (B) The three components, dipole voltage and two membrane
capacitative voltages, sketched separately. Different time constant 1: determine the rise
and fall of each component. A. Sketch of the sum of components, i.e., the total Open-
circuit photovoltage, under the condition 11 < r, < rm. Waveform (A) is the sum of (B)
on the same scale.

For this case, 1', r2, the waveform will look like Fig. 8A. Analysis of
the scope trace (Fig. 6) yields

1'1 = 30 :1: IO us
12 = 30 :l: Io ms (7)
r... = 840 ms

These values are comparable to those found by other methods?” with
the assurance that they have not changed in the lipid—aqueous solution
organized interface.

Under flash illumination, the decay of the photovoltage is due to
the depletion Of the excited molecules BR“. Under constant illumination,
when sufficient ionic charge has passed through the membrane tO com-
pensate the initial polarization charge Q, then V, is zero, and V,, is a
constant determined by the proton current. Removal Of illumination
quenches the dipole layer and leads to a dark-overshoot until charge
compensation again occurs. This effect is seen in other systems.3‘o37

522 Higgins, Lopez and Ti Tien

The potential difference across a uniform dipole layer is given by
the following equation

I .
V = :okm“ ./ 60'de (8)

in mks units, where so = 8.85><IO-m Farad/meter, k is the dielectric
constant Of the medium between the double charge layers that consti-
tute the dipole layer, e = I.60><IO“19 Coulomb (the proton’s charge),
a is the number Of deprotonated PM per m”, S is the mean separation
between the proton and PM‘, and e) is the solid angle subtended at the
electrode by the planar membrane. Let the membrane diameter be 1 mm,
the electrodes each be I cm away from the membrane, and S = 100 A,
and BLM dielectric constant k = 2. Then

a/V : IO“ A“in mVrl, (9)
and the average separation between the PM' and H+ is
5(A) z 10-3 [lemV)]-Vz. (10)

A typical Vd of our system is 25 mV, so 5 is 200 A. This small separation
is consistent with the rapid rise Of V, and with the diameters Of our
liposomes.

Summarizing, the existence of the dipole voltage, V4, is demonstrated
by the fast rise time Of the flash photovoltage, VP, and the fact that the
equivalent circuit (Fig. 7A) does not completely explain its waveform
(Fig. 6). (This circuit does completely explain Vp(t) under near short
circuit conditions, when the dipole field is shorted out. Then the shunting
resistor permits much faster charge compensation). The rise time (much
faster than I/rm) suggests rapid polarization Of the membrane by the
generation Of an electrical double layer inherent in the translocation Of
protons, and possibly the creation of temporary conformations Of PM
with electric dipole moment. The potential generated by the proton
current through the membrane is also present. Thus from equation 2,

Vp(t) = no) + V, (n)

where each term has different time constants. The proper circuit condi-
tions can make one term dominant at any given time. Recognition Of
this resolution Of photovoltage into two components is potentially im-
portant as a way to measure rate constants in the BLM environment, and
as a consideration for future study Of PM—BLM mechanisms.

H. halobium I. In vitro Studies 523

Acknowledgements

We thank Dr. JOSEPH Y. CASSIM, Ohio State University, Columbus,

for a sample Of H. halobium (R1). We are grateful to Dr. DONALD S.
BERNS, Department Of Health, State Of New York, Albany, for his gen—
erous gifts Of Halobacterium cells and purple membranes used initially
in this study. Financial support was provided by a grant from the U.S.
Public Health Service (GM—14971).

References

1
2

10

ll

12

18

14
15

16

17

18

19

20
21
22
23

H. LARSEN, Adv. Microb. Physiol. 1 (1967) 97

D. OESTERHELI and W. SrOEchNIUs, Proc. Natl. Acad. Sci. U.S.A. 70
(1973) 2853

P. MITCHELL, Biol. Rev. 4| (1966) 445

D. OESTERHELT, Angew. Chem. Int. Ed. Engl. I5 (1976) 17

TH. SCHRECKENBACH, in Photosynthesis in Relation to Model Systems, J. BAR-
BER (Editor), Elsevier—North Holand, Amsterdam, New York (1979) p. 189
R.A. BOGOMOLNI, L. 81013135 and J.K. LANYI, Biochemistry 11 (1978) 1037
J.B. HURLEY, T.G. EBREY, B. HONIG and M. OTTOLENGHI (1977), Nature
(London) 270 (1977) 540

E.P. IPPEN, C.V. SHANK, A. LEWIS and M.A. MARCUS, Science 200 (1978) 1279
MB. JACKSON and J.M. STUR'I‘EVANT, Biochemistry 17 (1978) 911

D. OESTERHELT, Ciba Found. Symp. Energy Transform. Biol. Systems 31
(1975) I47

M. YOSHIDA, N. SONE, H. HIRATA, Y. KAOAWA, Y. TAKEUCHI and K. OHNO,
Biochem. Biophys. Res. Comnmn. 67 (1975) 1295

E.P. BARKER, H. ROTTENHERG and SR. CAPLAN, Biochim. Biophys. Acta
440 (I976) 557

R.A. BOGOMOLNI, R.A. BAKER, R.H. LOZIER and W. S‘I‘OECKENIUS, Biochim.
Biophys. Acta 440 (1976) ()8

A. DANON and SR. CAPLAN, Biochim. Biophys. Acta 423 (1976) 133

Y. KAOAWA. K. OHNO, M. YosmnA, Y. TAHEUCHI and N. SONE, Fed. Proc.
Fed. Am. Soc. Exp. Biol. 36 (1977) 1815

N. SONE, Y. TAKEUCHI, M. YOSHIDA and K. OHNO, ]. Biochem. 82 (1977)

1751
C.D. WINOEI, N. KANNER and E. RACKIC, Biochim. Biophys. Acta 460
(I977)490

Z. DANCSHsz and B. KARVALY, FEBS Lett. 72 (1970') I36

L.A. DRACHEV, U..\'. FRULOV, A.D. KAULEN, EA. LIBERMAH, SA. OSTROU-
MOV, V.K. PLAKUNOVA, A.Y. SEMENOR and V.P. SKULACHEV, j. Biol. Chem.
251 (1976) 7059 .

P. SHIEH and L. PACKER, Biochem. Biophys. Res. Commun. 7| (1976) 603
M.C. BLOK, K.J. HELLINCWERE and K. VAN DAM, FEBS Lett. 76 (1977) 45
B. KARVALY and Z. DANCSHAZY, FEBS Lett. 76 (1977) 36

M.C. BLOK and K. VAN DAM, Biochim. Biophys. Acta 507 (1978) 48

524 Higgins, Lopez and Ti Tien

3‘ T.R. HERRMANN and G.W. RAYFIELD, Biophys. ]. 21 (1978) 111

25 H.T. TIEN, Bioelectrochem. Bioenerg. 5 (1978) 318

2“ H.T. TIEN and W. STILLWELL, s. Bioelectrochem. Bioenerg. following paper

27 B. BECHER and J.Y. CASSIM, Prep. Biochem. 5 (1975) 161

’3 H.T. TIEN, Bilayer Lipid Membranes (BLI‘W): Theory and Practice, Dekker,

New York (1974)

V.A. SINESHCHEKOV and F.F. erer, Biofizika 21 (1976) 313

H.T. TIEN and S.P. VERMA, Nature (London) 227 (1970) 1232

N. KOBAMOTO and H.T. 'TIEN, Biochim. Biophys. Acta 241 '1971) 129

32 RM. LOCSOM and H.T. TIEN, Chem. Phys. Lipids 8 (1972) 221

33 A. ILANI and BS. BERNS, ]. Membr. Biol. 8 (1972) 333

3‘ A.M. ULLRICH and H. KUHN, Biochim. Biophys. Acta 266 (1972) 584

35 H.W. TRISSL and P. LAUGER, Biochim. Biophys. Acta 282 (1972) 40

3‘ F.T. HONO and D. MAUZERALL, Proc. Natl. Acad. Sci. U.S.A. 71 (1974) 1564

3" J.S. HUEBNER, ]. IVIembr. Biol. 39 (1978) 97

3‘ K.J. KAUFMANN, V. SUNDSTROM, T. YAMANE and PM. RETNZEPIS, Bio-
phys. j. 22 (1978) 121

3’ R.H. LOZIER and W. NIEDERBERGER, Fed. Proc. Fed. Am. Soc. Exp. Biol.
36 (I977) 1805

388

APPENDIX C

APPENDIX C

Journal of Photochemistry, 20 (1982) 139 - 151 139

PHOTOELECTRIC EFFECTS IN BILAYER LIPID MEMBRANES
CONTAINING COVALENTLY LINKED PORPHYRIN COMPLEXES

N. B. JOSHI, J. R. LOPEZ and H. T. TIEN
Department of Biophysics, Michigan State University, East Lansing, MI 48824 (U.S.A.)

CHING-BORE WANG
Chemistry Department, Michigan State University, East Lansing, MI 48824 (U.S.A.)

QI-YI LIU
Physics Department, Nanhai University, Tainjin (China)

(Received March 15, 1982)

Summary

Complexes of retinal, carotene and quinone covalently linked to por-
phyrin were incorporated into bilayer lipid membranes (BLMs). The photo-
voltages and photocurrents in the BLMs were measured on continuous and
flash (8 us) illumination. The photovoltages obtained for tetraphenylpor-
phyrin-carotene and porphyrin-quinone membranes are much larger than
those obtained with simple porphyrin membranes. No pigmented BLMs
exhibit any latency in their photopotentials when illuminated with a light
flash of microsecond duration. The photovoltage and photocurrent action
spectra were measured for these pigmented BLMS. The photovoltage and
photocurrent observed for the pigmented membranes are explained in terms
of a semiconductor model. The enhancement in these photoeffects is
explained in terms of intramolecular charge transfer and energy transfer pro-
cesses.

 

l . Introduction

The formation of pigmented bilayer lipid membranes (BLMS) of planar
and spherical configuration [1, 2] has allowed a variety of photophysico-
chemical studies to be carried out to investigate the effects of chemical com-
position, redox compounds and the wavelength of light. The most important
feature of such studies may lie in the understanding they provide of the
transformation of solar energy into electricity or the storage of this energy in

0047-267 0/82/0000-0000/802J 5 © Elsevier Sequoia/Printed in Switzerland

171

140

chemical compounds as accomplished by green plant photosynthesis for
example [3 - 6]. Some of the advantages of the planar BLM system reside in
its relatively large size and its configuration which allows access to both its
sides and makes it relatively easy for electrical measurements to be carried
out. Studies on such pigmented membranes include investigations of mem-
branes formed from chloroplast extract and from purified chlorophylls and
their related compounds such as porphyrins with and without added
modifiers such as quinones and carotenes [2, 7 - 14] .

The photoeffects observed in such pigmented membranes are the result
of charge separation and electron transfer reactions (redox reactions) from
the excited pigment and suitable electron donors and acceptors at both inter-
faces. In photosynthetic reaction centers, charge separation and electron
transfer take place between the excited chlorophyll molecules and the
primary electron acceptors of the electron transport chain [3, 15]. There is
evidence that quinones play the role of the primary electron acceptors in
these reaction centers and that carotenes act as light-gathering accessory pig-
ments [3, 15]. For these processes to take place efficiently it seems reason-
able to assume that a relatively close proximity and proper orientation
should exist between the donor species (e.g. excited chlorophyll or por-
phyrin) and the electron acceptor (e.g. a quinone).

To test these hypotheses several groups of workers have synthesized
covalently linked porphyrin-quinone and porphyrin-carotene complexes
[16 - 20]. These investigators have studied the photophysical and photo-
chemical properties of these compounds in bulk solutions or on' surfaces.
Quenching of the porphyrin fluorescence in the presence of either quinones
or carotenes was observed. A similar effect was observed in chlorophyll
liposomes containing fi-carotenes [21] . The quenched fluorescence of
chlorophyll was ascribed to a static process of electron transfer or chloro-
phyll-chlorophyll interactions brought about by B-carotene.

The effect of quinone and fi-carotenes on some membrane model sys-
tems has also been investigated. Barsky et al. [22] have reported an increase
in photovoltages generated across a planar membrane containing reaction
centers from photosynthetic bacteria when 1,4-naphthaquinone is added to
the aqueous phase and when ubiquinone 30 is added to the membrane. The
presence of certain quinones in BLMS containing chlorophyll a has been
found to enhance both dark conductivity and photoconductivity [10].
Similarly, an increase in the photovoltage was found in the presence of [3-
carotene in the chlorophyll BLM [2] .

In all these previous studies dealing with the effect of quinones and
carotenes in pigmented BLM, however, compounds of porphyrins and/or
carotenes were either dissolved in the membrane-forming solution or added
to the aqueous phase, and the increase in the photoeffect was relatively
small. We report here the photoelectrochemical properties of BLMS contain-
ing retinal, carotene and quinone compounds covalently linked to porphyrin.
Preliminary results for quinone complexes have been communicated
previously [23].

141
2. Materials and methods

Anhydrous sodium acetate (AnalaR grade), FeCl3 (AnalaR grade;
Mallinckrodt Inc.), ascorbic acid, phosphatidylserine, phosphatidylcholine
and phosphatidylethanolamine (Sigma Chemical Co.) were used without
further purification. However, cholesterol (Fisher Scientific Co.) was
purified by repeated crystallization. The porphyrins with ester side chains
were prepared by the condensation of dipynomethene. The amines were
obtained from alcohols and mesylates [24]. The quinone compounds were
prepared following the published methods [16, 17]. However, in our case
silver oxide was used to oxidize the dihydroxy groups. Porphyrin-retinyl
was obtained by the condensation of porphyrin amine and retinal. All these
compounds were prepared under minimum illumination to protect them
from photodecomposition and were stored at a low temperature. The puri-
ties of the final compounds were checked by thin layer chromatography and
mass and optical spectroscopy. The tetraphenylporphyrin (TPP)-carotene
complex was a generous gift from Professor Gust, Arizona State University.
The synthesis and characterization of this complex has already been
published [19] . n-octane (Fisher Scientific Co.) and doubly distilled water
were used as solvents.

A solution of phosphatidylcholine, phosphatidylethanolamine,
cholesterol and phosphatidylserine in the ratio 3.1:1.7:1.1:1.0 by weight
was prepared in n-octane to a final concentration of approximately 7%. A
saturated solution of each porphyrin compound in the above solution was
used as the membrane-forming solution.

The membranes were formed by ejecting a small amount of membrane-
forming solution over the orifice (1 mm in diameter) of a 10 ml Teflon cup
separating two aqueous solutions (0.1 M sodium acetate buffer, pH 5.0)
using a Hamilton microsyringe. The Teflon cup was placed in a Plexiglas
chamber. The two faces of the Plexiglas chamber, one for illuminating and
the other for viewing the membrane, contained optically flat glass windows.
Calomel electrodes with salt bridges were used in the usual manner for elec-
trical contact [2] . The aqueous solutions were continuously agitated by
magnetic stirring bars in each of the chambers. The thinning of the mem-
brane to the secondary black state was observed with a binocular micro.
scope. FeCl3 and ascorbic acid were added to the inner and outer chambers
respectively to a final concentration of 10 mM after the membranes had
thinned and reached the black state.

The photopotentials and photocurrents were measured using a Keithley
610 electrometer and a Keithley 417 picoammeter connected to a Keithley
370 or Bausch and Lomb recorder. In some cases the signal was amplified
through a high impedance amplifier and recorded on a Tektronix R5031
dual-beam storage oscilloscope. The closed—circuit measurements were per-
formed by applying a potential through an external resistance which was
nearly equal to the membrane resistance in the dark. The membranes were
illuminated with white light from a 250 W tungsten-halogen lamp or a

142

1000 W xenon lamp. Microsecond flash (8 us) illumination experiments were
performed using a General Radio Strobotac 1538A xenon flash lamp assem-
bly. All the other experimental arrangements were the same as described
elsewhere [2] .

The experimental set-up for recording the action spectra consisted of a
1000 W xenon lamp assembly (Schoeffel), a Bausch and Lomb 250 mm grat-
ing monochromator, a pair of calomel electrodes, a Keithley 610 electrom-
eter and a Bausch and Lomb or Plotamatic 715M recorder. The spectra
were measured by rotating the diffraction grating with a slow speed motor or
manually point by point in increments of 10 nm. The action spectra were
corrected for constant illumination intensity with a calibration curve
obtained for the experimental arrangement using a Scientific Instrument
Kettering radiant power meter. The absorption spectra for BLM-forming
solutions (saturated solution of pigment) were recorded on a Varian Cary 219
spectrophotometer by sandwiching the solution between two glass plates. All
the experiments were performed at 23 t 1 °C.

3. Results

The membranes reached the black state 5 - 15 min after their forma-
tion. The membrane resistance Rm in the dark was measured by applying an
external voltage through a high resistance (R, z 108 $2) in series. The values
of Rat, for BLMs containing various porphyrin compounds are given in
Table 1. These values were measured after 3 - 4 h of membrane formation
when they were almost constant.

Open-circuit photo-e.m.f.s for various BLMs in the presence of FeCl3
and ascorbic acid in the aqueous solutions are given in Table 1. For most of
the compounds 8 small and slow photo-e.m.f. was observed initially; how-
ever, with time the photopotential showed an increase in magnitude and a
decrease in the response time and finally after 3 - 4 h it became almost
constant. This behavior was observed in BLMs containing porphyrin,
porphyrin-Cu and TPP-carotene. However, the response for the BLMs con-
taining porphyrin—quinone and porphyrin-retinyl was fast initially and
became much faster with time with a simultaneous increase in the magni-
tude. The rise and decay of the open-circuit pho_to-e.m.f. with continuous
illumination is shown in Fig. 1(a) for BLMs containing porphyrin,
porphyrin—quinone and porphyrin—retinyl. All the values of the open-circuit
photo-e.m.f.s given in Table 1 for BLMs containing various compounds were
recorded after 3 - 4 h of membrane formation when the photo~e.m.f. was
almost constant. The values of the photo-emf given in Table 1 clearly
indicate that it is enhanced in all BLMs containing monosubstituted com-
pounds except the BLM containing porphyrin-Cu. The BLMs containing
TPP—carotene and porphyrin-quinone showed more than threefold enhance-
ment compared with the BLM containing porphyrin. In BLMs containing
disubstituted complexes the photo-emf. was almost equal to or lower than
that of the BLM containing porphyrin.

TABLE 1

Photovoltage, photocurrent and membrane resistance

containing porphyrin complexes

143

for bilayer lipid membranes

 

 

 

Compound Photovoltags (mV) Photocurrent Rm 3

Continuous Flash unth continuous (x 10 $2)

. . . . . . illumination

illumination illumination

(11A)

Porphyrin 83 0.4 0.10 4
Porphyrin—retinal 154 20 1 .3 3
Porphyrin-diretinal 70 3.0 0.6 8
Porphyrin—quinone 302 42 22 8
Porphyrin—diquinone 92 1.0 1.2 3
Porphyrin—dodecyl 83 0.25 0.15 8
TPP-carotcnc 300 4.0 1 .2 1 1

 

The intensity of white light in continuous illumination was 250 mW cm“. The values of
the photo-e.m.f.s with flash illumination were measured 100 us after the flash. The struc-
tural details of the compounds are given in Appendix A.

L.
13
o‘

(b)

50 «W
0—0

 

l
'1: 3
57.:

Fig. 1. (a) Development of the photovoltage under continuous illumination with white
light for BLMs containing porphyrin (curve 1), porphyrin-quinone (curve 2) and
porphyrin-retinyl (curve 3) (I, light on; 1‘, light off); (b) development of the photovolt-
age under flash illumination (8 us duration) for BLMs containing porphyrin (curve 1),
porphyrin-quinone (curve 2) and porphyrin—retinyl (curve 3) (l, onset of flash illumina-
tion). The pigmented BLMs were formed in 0.1 M sodium acetate buffer (pH as 5.00)
with the outer solution containing 10 mM ascorbic acid and the inner solution containing
1 mM FeCl3. The intensity of the white light was 250 mW cm‘z. I

144

The membranes containing TPP-carotene, porphyrin-retinal and
porphyrin-quinone all exhibit a small photovoltage in the absence of a
donor or an acceptor in the bathing solution. The photopotential was found
to increase after the addition of FeCla to the inner chamber and showed
further enhancement after the addition of ascorbic acid to the outer
chamber. However, if only ascorbic acid was added to one side with no
FeCl3 on the other side, the photopotential observed was smaller than that
observed in the presence of FeCla only.

All photovoltage and photocurrent measurements were carried out in
the presence of 1 mM FeCl3 in the bathing solution in the inner chamber and
10 mM ascorbic acid in the bathing solution in the outer chamber. However,
for some membranes, the open-circuit photo-e.m.f.s were also measured for
different concentrations of FeCl,, while the ascorbic acid concentration was
kept constant (10 mM). The dependence of the photo-e.m.f. on the FeCl,
concentration for a BLM containing porphyrin-quinone is shown in Fig. 2.

Open-circuit photo-e.m.f. measurements were also carried out on BLMs
containing some of these compounds at higher pH values (sodium acetate
(0.1 M) buffer, pH z 7 .5) in the presence of FeCl; and ascorbic acid or of
ethylenediamine tetraacetic acid and anthraquinone-Z-sulfonate but the mag-
nitude of the photo-e.m.f. was lower at this pH. Our results are similar to
those reported by Jimbo et al. [25] for TPP-coated electrodes in the pre-
sence of some quinones. These workers have suggested that the protonation
of the porphyrin film surface plays an important role in the charge transfer
process.

Short-circuit photocurrent measurements with continuous light were
also made for BLMs containing porphyrin compounds in the presence of

 

300-

250 *

 

Photovolioge (mV)

 

 

 

Ol 02 0.3 0.4 0.5 0.6 07 0.8 09 LOO
FeCl, Concentration (mM)

Fig. 2. Dependence of the photovoltage on the FeCl; concentration in the inner chamber

of the bathing solution for a BLM containing the porphyrin-quinone complex. The bath-

ing solution was 0.1 M sodium acetate buffer (pl-i 2: 5.00) and the outer chamber con-
tained 10 mM ascorbic acid. The intensity of the white light was 250 mW cm‘z.

145

FeCl3 and ascorbic acid and the results are given in Table 1. For BLMs con-
taining porphyrin-quinone compounds the short-circuit current is about 200
times higher than for the BLM containing porphyrin.

The open-circuit photo-e.m.f. and the short-circuit photocurrent for the
BLMs were also measured for different intensities of white light. These
results are shown in Fig. 3 for the BLM containing porphyrin-quinone. Both
photovoltage and photocurrent are almost linear initially (i’. e. for low light
intensities) but the saturation point is reached much earlier for the photo-
voltage. Our results are similar to those previously observed for the chloro-
plast extract BLM and for photovoltaic cells [1, 2]. Similar experiments
were also performed for light of wavelength 400 and 500 nm using inter-
ference filters and the results are similar to those obtained for white light
illumination.

 

 

T T F I
_. 3
E -l8 5
" E
8. ~15 e
5.3 5
g «I2 ,8
'- 2
g - a.

‘3
l l l l

 

 

 

O 200 400 600 800

Light Intensity (mW/cm‘)

Fig. 3. Dependence of the photoeffects on the intensity of the illuminating light for a
BLM containing porphyrin-quinone: A, photovoltage; 0, photocurrent. The bathing solu-
tion was 0.1 M sodium acetate buffer (pH ~ 5.00) with 1 mM FeCl3 in the inner chamber
and 10 mM ascorbic acid in the outer chamber.

Apart from the continuous illumination, the open-circuit photovoltages
were also measured under flash illumination (8 us) for BLMs containing dif-
ferent porphyrin compounds and the photoresponses were observed. The
magnitudes of the photovoltage 100 us after the flash are listed in Table 1.
The development of the photovoltage for BLMs containing porphyrin,
porphyrin-quinone and porphyrin-retinyl is shown in Fig. 1(b). No delay
was found from the flash to the generation of the photovoltage, which was
limited by our 8 us flash unit. This behavior is similar to that observed earlier
for a chloroplast extract BLM [26].

The action spectra for the photoeffects of the porphyrin-containing
BLMs were obtained. In general these spectra follow the absorption spectra

.146

of the specific porphyrin complex'present in the membrane, Le. the posi-
tions of the maxima in the action spectra match the positions of the peaks in
the absorption spectra. Moreover, the positions of the maxima in the action
spectra and in the absorption spectra for the different porphyrin complexes
(including simple porphyrin) are almost identical. These results indicate that
the species responsible for the photoeffects is the porphyrin ring. However,
the relative intensity of the peaks in the action spectra for each porphyrin-
containing membrane is different from the relative intensity in their corres-
ponding absorption spectra. The difference is observed in the relative ratio of
the band at 400 nm (Soret band) to the bands in the lower energy region
beyond 450 nm (maxima at 500, 534, 574 and 624 nm). To investigate this
in more detail the photocurrent action spectrum for the porphyrin-quinone
membrane was measured (Fig. 4). A difference in the ratio of the bands in
the photocurrent spectrum compared with the absorption spectrum was also
observed. These results suggest that the quantum efficiency for the photo-
current may be different for the different absorption bands. A calculation of
the relative quantum efficiency at the different bands suggests that the
quantum efficiency at the lower energy bands is about twice the quantum
efficiency at 400 nm. A higher quantum efficiency for red light compared

 

 

   

 

 

ac . . . . . r 50
‘40
3
30 s
O .1 e-
3 §
3 s
g .5:
<1 ‘20 69:
4:0
.2-
300 ‘ 400 1 560 L 600 .. O

 

Wavelength (nm)

Fig. 4. Absorption spectrum of a saturated solution of the porphyrin-quinone-3 complex
in the BLM-forming solution sandwiched between two thin glass plates( ); photocur-
rent action spectrum of a BLM containing the porphyrin~quinone~3 complex corrected
for constant intensity of illumination ( ------- ).

 

147

With blue light has been reported for electrodes coated with manganese( III)
porphyrin [27] . One possible explanation for this difference may be the de-
crease in absorbance at the Soret band for higher pigment concentrations.
However, in our studies the absorption spectrum used for comparison is that
of a saturated solution. The only way to be certain would be to measure the
absorption speCtrum of the membrane itself. Owing to the thinness of the
BLM, this is a very difficult task. This is also the reason why we have not
measured the absolute quantum efficiency for any of the porphyrin com-
plexes.

4. Discussion

The photoelectric effects observed for BLMs containing porphyrin com-
plexes can be explained in the same way as for membranes made from
chloroplast extracts [2] . The observed photoelectric phenomena are due to
redox reactions occurring at both membrane-electrolyte interfaces with
electrons acting as the charge carriers across the membrane [1, 2, 7 - 14]. In
this respect the pigmented BLM is assumed to be an organic semiconductor
[2, 4] which is analogous to a Schottky barrier except that the BLM system
has two interfaces. Bearing this assumption in mind, one side of the mem-
brane acts as a photocathode (p-type junction) and the other side acts as a
photoanode (n-type junction) as shown in Fig. 5. The aqueous solution plays
the role of the metal at the BLM-electrolyte interface. The photovoltages
observed are influenced by the redox compounds (electron donors and
acceptors) present in the bathing solution and/or externally applied voltages.
It should be pointed out that current flow in the BLM is due to photogen-
erated and dissociated electron-hole pairs (excitons); however, unlike the
metal in a Schottky-type cell, the aqueous solution is not an electronic con~
ductor. Therefore redox reactions must take place at both interfaces in order
to complete the circuit. A more detailed description of the semiconductor
model of charge separation in BLMs has been discussed in detail elsewhere
[2, 28] .

The larger photoeffects observed in these experiments for the por-
phyrin complexes as compared with simple porphyrin can be explained in
terms of two processes taking place between the porphyrin and the covalent-
ly linked compound: (i) charge transfer and (ii) energy transfer. In the case
of the complexes consisting of benzoquinone molecules covalently linked to
a porphyrin molecule, the enhancement in the photoresponse is believed to
be due to the first process, i.e. intramolecular charge transfer between the
donor (porphyrin) and the acceptor (quinones). This intramolecular charge
transfer depends on the overlap between the highest filled donor orbitals and
the lowest filled acceptor orbitals and is determined by the interaction be-
tween the 1t electrons in the donor-acceptor pair. This interaction should
depend on the distance and orientation between the donor and acceptor
parts of the covalently linked complex. Evidence for this is given by the fact

148

To nieeeudng Instruments

   
 

Aqueous
solution

Aqueous
solution

Am —_
( Em he find“ (Donor)

 

Photoosinode Photoenode

 

 

 

 

 

 

 

Fig. 5. Schematic diagram and equivalent circuit of a pigmented BLM under illumination
with one membrane-electrolyte interface acting as the photocathode and the other as the
photoanode: c, calomel electrode; End”, standard redox potential; Rm, membrane dark
resistance; Cm, membrane capacitance; R9, membrane photoresistance; Cp, membrane
photocapacitance; Vp, membrane photopotential (see refs. 8, 12 and 27).

that the magnitude of the photoeffects varies for the quinones linked by dif-
ferent chain lengths to the porphyrin (see ref. 23). The largest enhancement
is observed for the quinone linked by the chain with the largest number of
atoms. The smaller photoresponse for the shorter chain may be due to im:
proper orientation between the two groups or too short a distance between
the two moieties. A smaller photoresponse for a closer proximity could be
due to a higher probability of the back reaction regenerating excited por-
phyrin and the oxidized acceptor competing strongly with the charge separa-
tion reaction. A possible explanation for the smaller photoresponse in the
double quinone complexes is related to the number of pigment molecules in
the membrane. It was observed that the solubility of the double quinone
complexes was much lower than that of the single quinone complexes. Our
results on the enhancement of the photoeffects in these quinone-porphyrin
complexes is further supported by some recent studies on fluorescence
quenching in similar types of complexes [16 - 20]. Quenching of the por-
phyrin fluorescence would indicate that there would be more excited mol-
ecules available for competing non-radiative processes including charge
transfer to take place. It should be mentioned, however, that the probability

149

for non-radiative processes may be increased more than that for charge
transfer since total quenching of fluorescence has been observed for a com-
plex containing four quinones covalently linked to a porphyrin [19] .

In the case of the covalently linked porphyrin-retinyl and TPP-
carotene the enhanced photoeffects can be explained in terms of intramo-
lecular energy transfer processes as well as intramolecular charge transfer pro-
cesses. Here we believe that the stacked conformation in which the it elec-
tron of the retinyl or carotene chromophore resides lies just abdve the mean
porphyrin plane. The retinyl or carotene moiety performs an antenna funcJ
tion by transferring the energy of the absorbed light to the singlet excited
state of the porphyrin moiety as was observed by Moore et al. [19] in their
optical and nuclear magnetic resonance bulk phase studies. These workers
have also observed a substantial quenching of porphyrin fluorescence by the
carotenoid moiety which suggests the possibility of the formation of a
charge-separated state and results in the enhanced photoeffect across the
BLMs in the present experiments. The quenching of chlorophyll fluorescence
by carotenoids was explained similarly by electron transfer from carotenoid
to chlorophyll by Beddard et al. [29]. Thus pigmented BLMs containing
porphyrin-retinyl or TPP—carotene complexes may be very useful from the
photosynthetic point of view because they can act both as antennae for
gathering light energy and as efficient charge separators. Again it is interest-
ing to note that the photoeffects in covalently linked porphyrin-diretinyl
complexes were smaller than in the single retinyl complexes. This can be
explained in the same way as in the case of covalently linked porphyrin-
diquinones.

Finally in the case of the porphyrin-dodecyl complexes, no enhance-
ment of the photoeffects was observed relative to simple porphyrin. The coo
valently linked moieties in this case cannot act as light-gathering entities nor
can they act as electron donors or acceptors. This is additional evidence in
support of our hypothesis of intramolecular charge transfer and energy
transfer in the quinone—porphyrin and retinyl-porphyrin complexes. We
propose that a porphyrin ring containing both covalently linked quinone and
B-carotene would probably generate a very large photovoltage and photo-
current. This type of compound may prove to be very useful in solar energy
conversion [3, 4, 6] .

Acknowledgments

We are grateful to Professor D. Gust, Chemistry Department, Arizona
State University, Tempe, for his generous gift of the TPP-carotene complex
used in these studies. This work was supported by National Institute of
Health Grant (SM-14971.

150

References

Gab“ NH

CD4

11

12
13
14
15

16
17
18
19

20

21

22

23

24
25

26
27
28
29

J. H. Fendler, J. Photochem.. 1 7 (1981) 303.

H. T. Tien, in J. Barber (ed.), Photosynthesis in Relation to Model Systems, North-
Holland, New York, 1979, pp. 116 - 173.

J. R. Bolton and D. 0. Hall, Annu. Rev. Energy, 4 (1979) 353 - 401.

H. Gerischer, Top. Appl. Phys, 31 (1979) 115 - 172.

M. Gratzel, Ber. Bunssnges. Phys. Chem., 84 (1980) 981 - 991.

A. A. Krasnovsky, A. N. Semenova and V. V. Nikandrov, Dohl. Aliad. Nauli S.S.S.R.,
262 (1982) 469. g

M. Mangel, D. S. Berna and A. Ilani, J. Membr. Biol., 20 (1975) 171 - 180.

F. T. Hong, in M. Blank (ed.), Bioelectrochemistry: Ions, Surfaces, Membranes.
American Chemical Society, Washington, DC, 1980, pp. 211 . 237.

S. 8. Brady and M. Brody, Photochem. Photobiol.. 26 (1977) 57 - 58.

D. Mauzerall, in H. Gerischer and J. J. Katz (eds.), Ligh t-i'nduced Charge Separation in
Biology and Chemistry, Dahlenlionferenzen, Verlag Chemie, Berlin, 1979, pp. 241 -
254.

S. W. Feldberg, G. H. Armen, J. A. Bell, C. K. Chang and C.-B. Wang, Biophys. J., 34
(1981) 149 - 163.

H.-W. Trissl, Biophys. J., 33 (1981) 233 - 242.

J. Kutnik and Z. Lozewska, Stud. Biophys.. 82 (1981) 127 - 135.

J. S. Huebner, Photochem. Photobiol. 35 (1982) 141.

J. Amesz and L. N. M. Duysens, in J. Barber (ed.), Pn'mary Processes of Photosynthe-
sis, Elsevier, Amsterdam, 1977, pp. 149 - 185.

T.-F. Ho, A. R. McIntosh and J. R. Bolton, Nature (London). 286 (1980) 254.

I. Tabushi, N. Koga and M. Yakagita, Tetrahedron Lett., 3 (1979) 257 - 260.

J. Dalton and L. R. Milgrom, J. Chem. Soc., Chem. Commun., (1979) 609.

A. L. Moore, G. Dirks, D. Gust and T. A. Moore, Photochem. Photobiol., 32 (1980)
691 - 695.

S. Nishitani, N. Kurata, Y. Sakata, S. Misumi, M. Migita, T. Okada and N. Natage,
Tetrahedron Lett., (22) (1981) 2099 - 2102.

M. Fragata, J. Colloid Interface Sci, 66 (1978) 470 - 477.

E. L. Barsky, Z. Dancshazy, L. A. Drachev, M. D. Ilina, A. A. Jasaits, A. A.
Kondrashin, V. D. Samuilor and V. P. Skulachev, J. Biol. Chem, 251 (1976) 7066 .
7071.

N. B. Joshi, J. R. Lopez, C.-B. Wang, Q.-Y. Liu and H. T. Tien, Biophys. J., 37 (1982)
228a.

C. K. Chang, J. Am. Chem. Soc., 99 (1976) 2819.

H. Jimbo, H. Yoneyama and H. Tamura, Photochem. Photobiol.. 32 (1980) 319 -
326.

J. S. Huebner and H. T. Tien, Biochim. Biophys. Acta, 256 (1972) 300 - 306.

Y. Umezawa and T. J. Yamamwa, J. Electrochem. Soc., 126 (1979) 705 - 707.

H. T. Tien, Sep. Sci. Technol, 15 (1980) 1035 - 1058.

S. G. Beddard, R. 8. Davidson and K. R. Trethewey, Nature (London), 267 (1977)
373 - 374.

151

"7 R'i2 Re

Porphyrm: R] E R3 5 R5 5 R7 5 CH3; R2 5 R6 5 CsH“; R4 5 R8 5
(CH2)2COOCH3; R9 5 R10 5 Ru 5 R12 5 H-
Porphynn-l‘etlnal: R1 5 R4 5 R5 5 R7 5 CH3; R2 5 R3 5 R6 '3 C5H11;R8 E
(Cfizlz—Cooczoflzei R9 5 R10 5 R11 5 R12 5 H-
Porphyrin-diretinal: R, E R3 E R5 5 R., s CH3; R2 2 R6 :‘—= CSH”; R., 5
Rs 5 (CHzlz-Cooczoflzsii R9 5 Rio 5 Ru 5 Ru 5 H-
Porphyrin-quinone: R1 5 R4 5 Rs 5 R7 5 CH3; R2 5 R3 5 R6 5 051-1“;
Rs 5 (CH2)3-N(Bu)—CO-C=C—C6H302; R9 5 Rio 5 R11 5 R12 5 H-
Porphyrin-diquinone: R, E R3 5 R5 5 R7 5 CH3; R2 5 R6 5 CSH”;
R4 5 R8 5 (CH2)3-N(BU)-CO—C=C_CGH302, R9 5 R405 R“ 5 R125 H.
Porphyrin dOdeCYI: R1 5 R4 5 R5 -=- R7 5 CH3;R2 E R3 5 R6 5 05H”;
Rs 5 (CH2)2—CONHC,2H25; R9 5 Rlo 5 R11 5 R12 5 H-
TPP‘CarOtene: R‘ E R25 R35 R45 R55 R65 R.7E R85 H; R.9E R10

O-carotene

Photobiochemistry and Photobiophysics, 4 (1982) 177-184 177
Elsevier Biomedical Press

PHOTOELECTROCHEMICAL PROPERTIES OF BILAYER LIPID
MEMBRANES CONTAINING COVALENTLY LINKED PORPHYRIN-
QUINONE AND OTHER COMPLEXES

CHING-BORE WANG, H. TI TIEN *, J.R.. LOPEZ, QI-YI LIU "‘, N .B. JOSHI and
Q.-Y. HU

Chemistry Department, Michigan State University, East Lansing, MI 48824. U.S.A.

Received February 11, 1982
Accepted February 14, 1982

 

SUMMARY

Several examples of a new class of covalently linked porphyrin-quinone
complexes have been incorporated into planar bilayer lipid membranes
(BLM) and shown to exhibit light-induced charge separation as demonstrated
by photoelectrospectrometry and electrical measurements. The photo-
electric effects obtained on some of these new BLM systems are three times
larger - the highest values hitherto reported - than those of chlorophyll-
containing BLMs.

 

INTRODUCTION

To elucidate the mechanisms of quantum conversion in algal and green
plant photosynthesis as well as to mimic Nature’s photochemical system for
possible solar energy utilization, a variety of model membranes has been
investigated [1-3]. One of the extensively studied model systems has been
the pigmented bilayer lipid membrane [4—12] of planar configuration,
commonly referred to as BLM, whose usefulness lies in its being an appro-
priate analogue of the thylakoid membrane [8,13]. For eflicient charge
separation and electron transfer in the thylakoid membrane, a close proxi-
mity between a donor species (e .g. chlorophyll or porphyrin) and an elec-
tron acceptor molecule (e.g. quinone) is a prerequisite. To test this hypo-

 

‘ To whom correspondence and requests for reprints should be addressed.
" On leave from Physics Department, Nankai University, Tainjin, China; Department of
Biophysics, Michigan State University, East Lansing, MI 48824, U.S.A.

0165-8646/82/0000—0000/$02.75 © 1982 Elsevier Biomedical Press

 

178

thesis several groups of workers have synthesized covalently linked por-
phyrin—quinone and porphyrin-carotene complexes [14—19]. We report
here the photoelectrochemical properties of BLMs containing quinones
and other compounds covalently linked to porphyrins. Our results support
the idea that covalently linked porphyrin-quinone compounds are more
efficient in charge separation and electron transfer than simple mixtures.
Previously, pigmented BLMs were formed from chloroplast extracts,
purified chlorophylls and their related compounds such as porphyrins
with and without added modifiers such as quinones and carotenes [8] .
The addition of quinones and/or carotenes is owing to the well-documented
evidence that quinones serve as the primary electron acceptor and caro-
tenes as effective lightgathering accessory pigments [20]. Incorporation
of the newly available porphyrin—quinone compounds (see Fig. 1), as a
model for the initial photophysicochemical event in reaction centers of
photosynthesis, into artificial BLMs is of obvious interest. The photo-
emfs obtained, for example, on some of these new BLM systems are indeed
interesting in that they are larger by at least a factor of three - the highest
values hitherto reported - than those of chlorOphyll-containing BLMs.

EXPERIMENTAL

Pigmented BLMs were formed by our standard technique [4] at room
temperature (23 r 1°C) in which a 5 pl of membrane-forming solution is
injected via a Hamilton syringe (Model PB-600) onto an aperture (diameter
:1 mm) in the wall of a 10 ml Teflon beaker separating two aqueous solu-
tions (0.1 M sodium acetate at pH 5). After the membrane formation, FeCla
(1 mM) on one side and ascorbic acid (10 mM) on the other side were added.
The membrane-forming solution consisted of a 3.1 : 1.7 : 1.1 : 1.0 mixture
(w/w) of phosphatidylcholine, phosphatidylethanolamine, cholesterol, and
phosphatidylserine, respectively, in n-octane saturated with different
porphyrin compounds. The different porphyrin compounds include: simple
porphyrin, and porphyrin with covalently linked quinone, retinal, choles-
terol, and dodecyl, etc. (Fig. 1). In the synthesis of the porphyrin—quinone
(PQ) complexes, we started with the coupling of porphyrin amino and cor-
responding acid chlorides. The PQ complexes were then obtained following
the published methods [14—19]. However, in our case, silver oxide was used
to oxidize the dihydroxy groups. The purity of the final compounds was
checked by TLC, mass and visible spectroscopy. Electrical prOperties of pig-
mented BLMs were measured using high impedance electrometers (Keithley
Instruments, Inc.). To observe photoinduced effects, light from either a
250 W tungsten-halogen lamp or a 1000 W xenon lamp was used. For flash
excitation a General Radio xenon lamp of 8 us duration was the light source
[8]. The action spectra of pigmented BLMs were obtained with the aid of a
Bausch and Lomb grating monochromator and a VOM chart recorder [22] .

W
‘s

1379

 

I. R1 - Ru - R5 -R7 - CH3, R2 - R3 - R6 - CSHll
A. R8 - 0 (CH2); - N(Bu) - C0 C6H302 (Q-l, P-BENZOOUINONYL)
E. R8 - - (CH2); - N(Bu) - C0 (CH2) C5H3)2 (0-2)
C. R8 ' - (CH2); - "(BU) ' C0 ‘ C : C -. C6H302 (0‘3)
D. R8 - - (CH2)2 - C00 CZOHZQ (VA,RETINYL)
E. R8 - - (CH2)2 - C00 C27Hq5 (CHOLESTEROL)
F. R8 . - (CH2)2 - COHH C12“2s (C12, DODECYL)
II. R1 ' R3 ' R5 - R7 ' CH3, R2 ' R6 ' CSHll
A. Ru ' R8 ' Q'l
B. Ru - R8 - 0-2
C. Ru - R8 - 0-3
D. Ru ' R8 ' 'CHZ ‘ C00 Conzg
E. Ru - R8 - -(CH2)2 ~ C00 C27Hu5
F. Ru - R8 - -(CH2)2 C00 CH3

[11. R1 ' R2 ' R3 ' Ru 'RS ‘ R5. C2 ”5’ R7 ‘ CH3
R8 - c(CH2)2 CONH (C9 H17 N0), Fae: RADICAL

IV. R1 ' R3 ’ Czfls, R2 ‘ Ru 'Rfi ' R8 ' CH3
R6 ' R7 ' - (CH2)2 ‘ CONH Clezs

Fig. 1. Structural formula and notations of compounds used.

RESULTS AND DISCUSSION

The effect of light on the pigmented BLMs can be easily detected by elec-
trical methods, which may be combined with optical techniques to yield
additional information concerning the pigmented BLM system [22]. The
open circuit photovoltage and short circuit photocurrent for BLM containing
different porphyrin compounds are given in Table I. The values were mea-
sured after 3-4 h of membrane formation when they were almost constant.
The results presented in Table I clearly indicate that quinone-substituted
complexes are most efficient in charge separation. Among the quinone-
substituted complexes the PQ-3 membrane shows nearly 3-fold increase in
photovoltage and ZOO-fold increase in photocurrent as compared to simple
porphyrin membrane. In disubstituted PQ complexes, however, the magni-
tude of the photoeffect was lower or almost equal to that observed for
simple porphyrin membrane. The time course of the photoresponse for PQ-3

180

TABLE I

TYPICAL RESULTS OF PIGMENTED BLM AS A FUNCTION OF PORPHYRIN COM-
PLEXES

 

 

Compound ‘ Photo-emf (mV) Photocurrent (nA) Rm (108 52)
(IV) Edi C-12 42 0.01 9
(II A) P-di Q-l 68 0.16 9
(II D) Pdi VA 70 0.6 8
(II B) P-diQ-2 71 0.06 11
(IF) P-C-12 83 0.15 8
(II F) Porphyrin 83 0.10 4
Chloroplast extract 86 0.75 5
(IE) P-Chol 88 0.15 5
(II C) P-diQ-3 92 1.2 3
(III) P-Spin 112 0.18 6
(I D) P-VA 154 1.2 3
(II E) P-diChol 175 0.84 6
(I B) PQ-2 190 12 4
(I A) PQ-l 194 15 7
(I C) PQ-3 302 22 8

 

‘ For abbreviations used see Fig. 1.

 

 
   
 

Chl-I LI

TPP
I- 50

 

 

1 n
‘ n In

)-
p

\\

i
i
t
s

 

DARK HEIIIANI PO‘I’INTML (mV)

Fig. 2. Effect of externally applied voltage on the photopotential of different pigmented
BLMs. The pigmented BLMs were formed in 0.1 M sodium acetate at pH 5 with outer
solution containing 10 mM ascorbate and inner solution containing 1 mM ferric chloride.
The external potential was applied under closed circuit conditions by means of a battery.
Each datum point represents the maximum potential developed at each value of exter-
nally applied potential when the BLM was illuminated. The illumination period for each
datum point was ~15 s. The maximum photopotential was reached in <10 s. See Table
I for abbreviations used.

181

membrane for two different intensities of illuminating light is given in Fig.
43. For PQ-3 membrane the photo-emfs were also measured for different
concentrations of FeCl, while keeping ascorbic acid concentration constant.
The magnitude of the photo-emf followed the stoichiometry of Fe” and
ascorbic acid.

In Fig. 2 the effect of an externally applied electric field on the photo-
response of five different pigmented BLMs is shown. The external field was
established by applying a potential difference by means of a battery under
closed circuit conditions. As can be seen from the figure, the presence of the
electric field has the same effect on the photoresponse for the five BLMs
containing different pigments. Depending on the polarity of the externally
applied field the photopotential can be either increased or decreased. The
greater the applied voltage, positive on the side containing the electron
acceptor (i.e. Fe”), the larger was the observed photopotential and vice
versa. The results can be explained in terms of the ultra-thinness of the BLM
(<10 nm) and its very high resistance (R z 108 $2 - cm" , several orders of
magnitude larger than the aqueous solution), which means that the applied

A

-I 0.0

 

 

oar-

 

 

 

 

a
m

WAVELENGTH (nm)

Fig. 3. A) Absorption spectrum of porphyrin-quinone complex PQ-3 in 1 : 10 methylene
chloride: n-octane solution (PQ-3 concentration: 10 mM). B) Photovoltage action spec-
trum for BLM containing PQ-3 with 0.5 mM FeCl3 on one side of the bathing solution,
0.1 M sodium acetate at pH 6. The spectrum has been corrected for incident energy. The
arrows indicate the y-axis for each spectrum.

182

voltage will produce a very high electric field across the BLM. This large field
strength (~10‘ volts/cm) will increase or decrease, depending on its polarity,
the. number of light-generated charges moving to Opposite sides of the BLM.
Fig. 3 shows the wavelength dependence of a PQ-3 containing BLM. The
action spectrum follows closely the absorption spectrum (curve A) of the
membrane-forming solution. This provides strong evidence that the photons
absorbed by the pigment only are responsible for the initial separation of
electronic charges in the BLM and the observed light-induced effects. To
obtain information on the fast photoprocesses and light-induced effects on
the pigmented BLM system, the membranes were excited by light pulses of
8 us duration. The photopotentials generated were recorded with different
time scales. The results for a BLM containing PQ-3 are shown on Fig. 4A.
When observed in this way, the photoresponse shows to be composed of
several components with different rate constants. Since the pigmented BLM
possesses an appreciable capacitance (Cm 2.- 5 X 10" fared) and resistance
(Rm 2 1—9 X 10‘ 52), the membrane is electrically equivalent to a capacitor
shunted with a resistor. The voltage transients appear across the effective

VA

W l 10ms

non-w _ 7

 

Vit immW/cllz mwxenZ
0

 

03,4.

 

‘l

Fig. 4. A) Time course of the photoresponse from a PQ3mntaining BLM excited by
short light flashes (8 us duration). The time course recorded with three different time
scales is shown. The downward arrows indicate the onset of the flash. B) Time course of
the photoresponse with continuous illumination for a PQ-3-containing BLM with two dif-
ferent light intensities in mW - cm" .

183

membrane Rm—Cm combination as previously described [5-10]. The initial
phase commences with about less than 1 us delay and grows during the flash
(Fig. 4A, upper left and middle curves). The final phase (Fig. 4A, right) is
the decay of the flash-induced voltage lasting several seconds. The fast
appearance of the photopotential across the membrane in the us range
(resolution limited by our apparatus) strongly suggests the generation and
separation of electronic charges. The continuing growth of the voltage after
the flash excitation implies the existence of long-lived species (e .g. triplet
state excitons or free radicals) which upon oxidation and reduction by elec-
tron acceptors and donors, respectively, in the aqueous solution contribute
further to the potential difference across the BLM.

A more detailed description of the possible sequence of events taking
place at both interfaces can be expressed in terms of the following reactions:

P-Q+hv-* [P—Q]**P’-Q' (1)
P’-Q‘ + A -> P’-Q + A' (2)
P*—Q‘ + D -> P—Q‘ + D’ (3)

P’—Q‘ is, as mentioned previously, a long-lived species whose nature is at
present not possible to identify exactly. The species P’-Q and P—Q' can
then be reduced and oxidized, respectively, either by electrons and/or holes
from donors (D) and acceptors (A) on opposite sides of the BLM in the bath-
ing solution. The net reactions for these two cases are:

P'—Q+D->P—Q+D. (4)
p—Q'+A->P—Q+A‘ (5)

Mechanistically, a pigmented BLM/aqueous interface has been likened to
that of a Schottky barrier [23], with the aqueous solution playing the role
of the metal. In this way the membrane is considered to be capable of elec-
tronic conduction.

From our results obtained using P—Q complexes, it is clear that the
enhanced photoeffect in the BLM system can be explained in terms of intra-
molecular charge-transfer processes between donor and acceptor moieties in
these covalently linked compounds. The only difference between the three
P—Q complexes is the chain length between the porphyrin and quinone
moiety (see Fig. 1). Therefore, we attribute the difference in the photo-
responses to the difference in the distance and orientation between the two
moieties. The interaction between the 1r electron in the two chromophores,
which depends on the distance and geometry of these chromophores in the
covalently linked complexes, should be maximum. It appears that by
covalently attaching these compounds, a more favorable orientation and
closer proximity is attained between the donor and acceptor pairs which
result in a maximum overlap between highly filled donor orbitals and lowest

184

filled acceptor orbitals. This provides more favorable conditions for eventual
charge separation rather than being dissipated by other pathways such as
fluorescence. In fact, a quenching of the fluorescence emission for porphy-
rin—quinone and other complexes has already been reported [15—19]. Our
results are consistent with those obtained by other workers on fluorescence
quenching. In the presence of such quenching the fluorescence quantum
yield will be low because of the utilization of all energy in charge separation
which results in net enhanced photoeffect. Thus, our findings provide sup-
port to the idea that a more efficient charge separation can take place when
the donor/acceptor pairs are in close proximity to each other. These cova-
lently linked compounds, in particular those typified by the PQ-3 type,
may be very useful in certain biomimetic solar energy transduction systems
[1—3,23,24] .

ACKNOWLEDGEMENT

This research was supported by grants from the NIH (GM-14971) and
DOE (DE-FG02-SOCS83101).

REFERENCES

Fendler, J.R. (1981) J. Photochem. I 7, 303-310.

Barber, J. (ed.) (1979) Photosynthesis in Relation to Model Systems. Elsevier, Am-

sterdam.

Bolton, J.R. and DD. Hall (1979) Ann. Rev. Energy 4. 353—401.

Tien, H.T. (1963) Nature 219, 272—274; J. Phys. Chem. 72, 4519-4521.

Trissl, H.W. and P. Linger (1970) Z. Naturforsch. 25. 1059-1061.

Ullrich, H.M. and H. Kuhn ( 197 2) Biochim. Biophys. Acta 266, 584—596.

Hong, F.T. (1977 ) J. Colloid Int. Sci. 58. 471-486.

Tien, H.T. (1976) Brookhaven Symp. Biol. 28, 105—131.

Huebner, J. (1979) Photochem. Photobiol. 30, 233—241.

Mauzerall, D. (1979) in: Light-induced Charge Separation in Biology and Chemistry

(Gerischer, H. and J.J. Kata, eds.) pp. 241-254. Dahlem-Konferenzen, Verlag

Chemie, Weinheim.

11 Alexandrowicz, G. and BS. Barns (1980) Photobiochem. Photobiophys. I. 353-360.

12 Feldberg, S.W., O.H. Armen, J.A. Bell, C.K. Chang and C.-B. Wang (1981) Biophys.
J. 34, 149-163.

13 Witt, H.T. and D. Difiore (1981) FEBS Lett. 128, 149-153.

14 Tabushi, 1., A. Koga and M. Yakagita (1979) Tetrahedron Lett. 3, 257-260.

15 Kong, J.L.Y. and P.A. Loach (1980) J. Heterocyclic Chem. 1 7, 737-744.

16 Ho, T.-F., A.R. McIntosh and J.R. Bolton (1980) Nature 286, 254—256.

17 Dalton, J. and L.R. Mila'om (1979) JCS Chem. Commun. 609—611.

18 Moore, A.L., G. Dirks, D. Gust and T.A. Moore (1980) Photochem. Photobiol. 32.
691—695. .

19 Nishitani, S., N. Kurata, Y. Sakata. S. Misumi, M. Migita, T. Okada and N. Nataga
(1981) Tetrahedron Lett. 22, 2099—2102.

20 Olson, J.M. and G. Hind (ed.) (197 7) in: Chlorophyll-Proteins, Reaction Centers, and
Photosynthetic Membranes. National Technical Information Service, Srpingfield, VA.

21 Metzner, H. (ed.) (1978) Photosynthetic Oxygen Evolution. Academic Press, Lon-
don-New York—San Francisco.

22 Lopez, J. and H.T. Tien (1980) Biochim. Biophys. Acta 597, 433—444.

23 Tien, H.T. (1980) Sep. Sci. Tech. 15, 1035-1058.

24 Kampas, F.J., K. Yamashia and J. Fajer (1980) Nature 284, 40—42.

NH

O'DCDQOQO‘IFW