THE EFFECT OF TEMPEfiKTURE on me
BIPHASIC PHOTOFOTENTIALS‘ OF ALL 5mm . RE‘HNAL
BIMOLECULAR um MEMBRANES

Thesis for the Degree of Ph. D.
MICHEGAN STATE LEMVERSETY
NAGTADA KGBASAOTO
1970

This is to certify that the
thesis entitled
THE EFFECT OF ”EJPERATURE ON THE BIPHASIC
PHD TOPO TENTIALS OF ALL-TRAJS-RETINAL
BIMO LECULAR LIPID I'lhiiuflfifli’illis

presented by

I'Iaotada. Kobasnoto

has been accepted towards fulfillment
of the requirements for

 

 

Date November 18, 1910

0-7639

 

LIBRARY

Michigan State
University

 

 

.3\17-—rf‘22;7 {g

ABSTRACT

THE EFFECT OF TEMPERATURE ON THE BIPHASIC
PHOTOPOTENTIALS OF ALL-TRANS-RETINAL
BIMOLECULAR LIPID MEMBRANES

BY

Naotada Kobamoto

The bimolecular lipid membranes formed from lipid
solutions containing all-transfretinal could generate,
upon illumination, temperature-sensitive biphasic photo-
potentials which were quite similar to the early
receptor potentials of visual receptors and pigmented
animal and plant tissues: (1) The photoresponses consist
of the fast component (R1), and the slow component (R2)
which showed the sign Opposite to R1. (2) At proper
experimental conditions a complete phase shift from R1
to R2 took place as the temperature increased from 9°C
to 45°C. (3) R1 and R2 were not influenced by changes
in the ionic strength of K+, Na+, and C1-, nor by
generation of concentration gradient of these ions across

the membrane.

Naotada Kobamoto

R1 was concluded to be the potential which was
mainly determined by hole diffusion from the following
observations: (1) The sign and magnitude of R1 could be
influenced only by strong electron acceptors such as
ferric ions. (2) R1 was independent of pH gradient and
buffer capacity. (3) The side containing ferric ions
showed negative photopotentials.

R2 was identified as the potential which was strongly
dependent on protonic diffusion by the following observa—
tions: (1) Changing proton concentration gradient across
the membrane could control R2 in such a manner that the
magnitude of R2 was linearly proportional to the logarithm
of the proton concentration gradient. (2) R2 was
abolished in the presence of strong buffer capacity.

The effect of increasing temperature could be seen
as an enhancement not only in R2 but also in dark membrane
potentials and membrane conductances. This observation
was interpreted to mean that the effect of temperature
was to increase proton mobility in the hydrocarbon layer
of the membrane. When this finding was applied to the
theory develOped in this work to account for the time
course of the photopotentials, the temperature dependence
of the biphasic photopotentials could be explained
satisfactorily as the consequence of an increase in the

proton conductance relative to the hole conductance.

Naotada Kobamoto

The mode of biphasic photopotential could also be
influenced by pH gradient across the membrane, the buffer
capacity of the aqueous solutions, and light intensity.
The theory developed in this work could account for these
observations as well as the temperature effect.

A possibility of applying the observations carried
out in this work to the elucidation of the mechanism of

generation of the early receptor potentials was examined.

THE EFFECT OF TEMPERATURE ON THE BIPHASIC
PHOTOPOTENTIALS OF ALL-TRANS-RETINAL

BIMOLECULAR LIPID MEMBRANES

BY

Naotada Kobamoto

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Biophysics

1970

ACKNOWLEDGMENTS

The writer wishes to express his thanks to Professor
H. Ti Tien for suggesting this investigation as well as
for constant guidance in the course of this work.
Especially, it is due to Professor Tien's deep insight
into the nature of bimolecular lipid membranes that the
successful formation and characterization of the membrane
system used in this work have been achieved.

Thanks are also due to Dr. Surendra Verma, Dr. Hie
Ping Ting, Dr. Nguen Thuong Van, Dr. Fred M. Loxsum, and
Mr. William A. Huemoeller for the invaluable discussions
rendered during the course of this work, and to Drs.
Barnett Rosenberg and Harish C. Pant for critical reading
of the thesis.

Thanks are extended to the National Institutes of

Health for supporting this work with Grant GM-l497l.

ii

TABLE OF CONTENTS

ACKNOWLEDGMENTS . . . . . . . . . . . .
LIST OF TABLES . . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . . . . . .
Chapter
I. INTRODUCTION . . . . . . . . . . .
II. LITERATURE REVIEW . . . . . . . . .

III.

IV.

Early Receptor Potential . . . . . . .
The Effect of Temperature on the Early
Receptor Potential . . . . . . .
The Necessity of the Membranous Structure for
the Generation of the Early Receptor
Potential . . . . . . . . .
The Necessity of the Mediator for the Genera-
tion of the Early Receptor Potential . .
The Selection of the Membrane Model System .
The Importance of Membrane Models in the
Studies on the Early Receptor Potential .
The Structure of the Membranes of Rods
and Cones . . . . . . . . . . .
Photosensitive Monolayers . . . .
Photosensitive Bimolecular Lipid Membrane .

THE CONSTRUCTION OF THE MODEL MEMBRANE
SYSTEM USED IN THE PRESENT WORK . . . . .

All-Trans-Retinal Bimolecular Lipid
MefiBrane . . . . . . . . . . . .
Aqueous Phases . . . . . . . . . .

EXPERIMENTAL . . . . . . . . . . .

Lipid Solutions . . . . . . . . . .
Membrane Forming Solution . . . . . . .
Chemicals . . . . . . . . . . . .
Aqueous Solutions . . . . . . . . .
Membrane Formation . . . . . . . . .

iii

Page
ii
vi

vii

11

‘13

15
21

21

24
25
27

29

29
33

35

35
36
36
37
37

Chapter Page

Potential Measurements . . . . . . . . 40
Resistance Measurements . . . . . . . . 41
pH Measurements and Control . . . . . . 44
Temperature Measurements and Control . . . 44
Light Sources . . . . . . . . . . . 44
Filters . . . . . . . . . . . . . 45
Artifacts . . . . . . . . . . . . 45
Transference Numbers . . . . . . . . . 46
V. RESULTS . . . . . . . . . . . . . 48
Dark Membrane Potentials .. . . . . . . 48
The General Description of the Photoresponses. 54
Light Intensity Dependence of the Photo-
responses . . . . . . . . . 70
The Effect of Temperature on the Photo-
responses . . . . . . . . 88
The Effect of pH on the Photoresponses . . . 110
The Effect of Buffer Capacity on the Photo-
responses . . . . . . . . . . . . 110
VI. DISCUSSION . . . . . . . . . . . . 126
General Discussion . . . . . . . . . 126
Theory . . . . . . . . . 131
The Effect of Temperature on the Biphasic
Photopotentials . . . . . . . . 160
The Effect of Temperature on the Dark
Membrane Potentials . . . . . . 160
The Effect of Temperature on the Membrane
Conductivity . . . . . . 165
The Effect of Temperature on the Photo-
responses . . . . . . . . . . 172
The Thermal Coefficients of Photoresponses . 184
The Further Experimental Verification of the
Theory . . . . . . . . . . . 193
The Effect of Light Intensity: Logarithmic
Dependence . . . . . . . . 193
The Effect of the Outside pH . . . . . 198
The Effect of Buffer Capacity . . . . . 200
The Effect of Light Intensity: Threshold
Requirement . . . . . . . . . . 201
"Hole-Proton Diffusion" Mechanism and Its
Alternatives . . . . . . . . . . . 203

VII. TOWARD A FURTHER UNDERSTANDING OF EARLY
RECEPTOR POTENTIALS . . . . . . . . . 205

The Membranous Origin of the Early Receptor
Potentials . . . . . . . . . . . 206

iv

Chapter Page

The Coupling of Electronic Transport with

Ionic Transport in Visual Excitation . . . 210
The Photogeneration of Protons in Photo-
receptors . . . . . . . . . . 212
The Protonic Induction of Ionic Membrane
Excitation . . . . . . . . . . 217
The Effect of Temperature on Proton Activity
in the Induction of Photopotentials . . . 221
VIII. SUMMARY . . . . . . . . . . . . . 225
BIBLIOGRAPHY . . . . . . . . . . . . . . 227

LIST OF TABLES

Page

Cationic transference numbers of all-trans-
retinal BLM o o o o o o o o o o o o o 51

The development of the dark membrane potential as
a function of the buffer concentration in the
aqueous phases . . . . . . . . . . . . 53

The effect of temperature on the duration of
illumination for the appearance of R2 and the
disappearance of R1 . . . . . . . . . . 62

The thermal coefficients of hydrogen ion transfer

number, membrane resistance, and R1 and R2
components . . . . . . . . . . . . . 103

vi

Figure

10.

11.

12.

13.

LIST OF FIGURES

Page

The schematic diagrams of the photoreceptor cells
from which the early receptor potential has been
recorded . . . . . . . . . . . . . 19

The apparatus for the formation of BLM and the
measurement of membrane potentials . . . . 38

Circuit diagram for the measurements of
resistance . . . . . . . . . . . . 42

Cation induced potentials as a function of
concentration . . . . . . . . . . . 49

The effect of the position of FeCl3 with
respect to the incident light on the mode of
photoresponse . . . . . . . . . . . 55

The effect of KCl concentration gradient across
bimolecular lipid membrane on the photoresponse 57

The effect of the duration of illumination on
the mode of photoresponses . . . . . . . 60

The effect of repetitive illuminations on the
photoresponses . . . . . . . . . . . 63

Photopotential (R2) and dark membrane potential
as a function of time from the first illumina-
tion . . . . . . . . . . . . . . 66

The ratio of photopotential (R2) to the dark
membrane potential as a function of the time
from the first illumination . . . . . . . 68

The effect of relative intensity of light on Rl
component at ApH 0.3 as a function of tempera-
ture O O O O O O I O O O I O C O 71

The effect of relative intensity of light on R1
component at ApH 0.6 as a function of tempera-
ture O O O O O O O O O I O O O O 73

The effect of relative intensity of light on R1
component at ApH 1.0 as a function of tempera-
ture . . . . . . . . . . . . . . 75

vii

Figure

14.

15.

l6.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

The effect of relative intensity of light on Rl
component at ApH 2.0 as a function of tempera-
ture I O O O O I O O O C O O O I

The effect of relative intensity of light on
R2/Rl at ApH 0.3 as a function of temperature .

The effect of relative intensity of light on
R2/R1 at ApH 0.6 as a function of temperature .

The effect of relative intensity of light on
R2/Rl at ApH 1.0 as a function of temperature .

The effect of relative intensity of light on
R2/Rl at ApH 2.0 as a function of temperature .

The effect of light intensity on the photo-
potential of all-trans-retinal bimolecular
lipid membrane . . . . . . . . . . .

The time course of the photoresponse of all-
trans-retinal BLM . . . . . . . . . .

The ratio of R2 to R1 as a function of
temperature . . . . . . . . . . . .

The time course of the photoresponses of
all-trans-retinal BLM as a function of
temperature: HCl system . . . . . . . .

The time course of the photoresponses of a11-
trans-retinal BLM as a function of temperature:
FeCI3 system . . . . . . . . . . . .

The temperature dependence of the dark membrane
potential of all-trans-retinal BLM in asym-

metric HCl solution or asymmetric FeC13
solution . . . . . . . . . . . . .

Dark membrane resistance and R1 photopotential
as a function of temperature in HCl system . .

Dark membrane resistance and R1 photopotential
as a function of temperature in FeC13 system .

The relationship between two components of
photopotentials and dark membrane resistance .

The effect of pH on the photoresponses . . .

R1 and R2 as a function of pH gradient across
the membrane . . . . . . . . . . . .

viii

Page

77

79

81

83

85

89

91

94

96

95

101

104

106

108

111

113

Figure Page

30. The time course of the photoresponses of a11-
trans-retinal BLM in symmetric KCl solutions
as a function of salt concentration . . . . 115

31. The time course of the photoresponses of all-
trans-retinal BLM in symmetric acetate buffer
solutions as a function of buffer concentra-
tion . . . . . . . . . . . . . . 118

32. The magnitude of photoresponses and dark
membrane resistance as a function of the con-
centration of KCl or sodium acetate buffer
systems . . . . . . . . . . . . . 120

33. The time course of photoresponses with 60
second illuminations at pH 3.3 and ApH 0 . . 123

34. The equivalent circuits for the generation of
photoresponses of all-trans-retinal BLM . . . 135

35. The analysis of two components of photo-
potentials . . . . . . . . . . . . 139

36. The photochemical reactions and diffusion
processes involving excitons (e), holes (h),
and hydrogen ions (H) . . . . . . . . . 149

37. Simplified photochemical reactions and dif-
fusion processes . . . . . . . . . . 152

38. Possible redox reactions at the membrane
surfaces . . . . . . . . . . . . . 155

39. The analysis of the time course of the biphasic
photoresponses of all-trans-retinal BLM . . . 176

 

40. The electronic energy levels of all-trans-
retinal bimolecular lipid membrane . . . . 190

41. The schematic diagram of photoelectric trans-
duction showing how hydrogen ion generation
and transport can function as the intermediate
stages (early receptor potentials) leading to
the ultimate Na+ and K+ transport (the earlier
component of electroretinogram) across the
photoreceptor membrane . . . . . . . . 218

ix

CHAPTER I

INTRODUCTION

Since the first report on the techniques of the
formation and properties of black lipid membranes in
aqueous solutions by Mueller, Rudin, Tien, and Wescott
(1962a, b, 1963, 1964), quite a new area of membranology
has been developed. The revival of colloid chemistry was
inevitable, since the structure, stability, and formation
processes of this ultra-thin membrane had to be understood
in terms of interfacial thermodynamics and molecular inter-
action between polar or non—polar portions of surface
active lipid molecules. Indeed, at the time of the first
publication, the above workers knew not only most of the
basic properties of this new type of membrane but also the
biological implication of their model membrane as the
manifestation of the Gorter and Grendel (1925) theoretical
model of the plasma membrane of red cells, which was then
known as the unit membrane model generalized by Danielli
and Davson (1935). In less than one decade this model
membrane provided an enormous amount of information in the
field of colloid science and biological membranology.

The fundamental importance of the Mueller-Rudin-Tien-
Wescott model membrane is that it can have stability for

1

hours in aqueous solutions. This fact is a strong support
for the Gorter-Grendel-Danielli—Davson membrane model,
in which the hydrophobic-interior and hydrophylic-exterior
type molecular arrangement was proposed to have enough
strength to function as a cellular barrier, since the
electron microsc0pic, Optical, and electrical determina-
tions of the membrane thickness and the molecular arrange-
ment of phospholipids in this membrane now leave no doubt
on the bimolecular leaflet nature of this model membrane.
This in vitro formation of cellular membrane, i.e.
Mueller-Rudin-Tien-Wescott model membrane, can be
achieved by painting a hole or ring (about 3 mm diameter
hole in Teflon or polyethylene support) with a minute
amount of biological or abiological surfactants dissolved
in normal alkanes of carbon length from six to fourteen.
If the surfactant concentration is proper, the thinning
process takes place immediately and a completely black
membrane surrounded by the Plateau-Gibbs border (the
lipid reservoir) can be formed within minutes. When this
thinning process is viewed under a proper arrangement of
light source and a low power (about 20x) microscope, vivid
and bright colorful fringes can be seen. An initially
colored film thins down by drainage of the lipid to the
border. After the colored film turns to a bright silver,
the nucleation of the black spot appears, usually near the

border of the silver film. It pushes away the colored

portion toward the Opposite border region, and finally
covers the total film area. This state of the lipid film
has been called a "black lipid membrane," "bilayer lipid

membrane," or "bimolecular lipid membrane,‘ abbreviated
quite often as BLM (Tien and Diana, 1968).

Excellent review articles in the general field of
BLM are now available. The works up to 1967 covering the
techniques of membrane formation and the basic physico—
chemical prOperties were extensively reviewed by Tien
and Diana (1968). Similar review was presented by Bangham
(1968) in conjunction with various other membrane models.
Rothfield and Finkelstein (1968) covered the works on the
membrane model from the view point of membrane biochemistry.
Reeves (1969) and Castleden (1969) reviewed the value of
the BLM as a model of the biological membrane in compari-
son with other membrane models. The general aspect of
this field from the point of View of the biological
application Of BLM were reviewed by Henn and Thompson
(1969). A recent review by Goldup, Ohki, and Danielli
(1970) is comprehensive and fascinating in that one of
the proponents of the bilayer membrane structure examines
the progress made in the field.

A recent unique application of the BLM system is
successful formation of photosensitive membranes by
incorporating commercial chlorophyll, xanthophyll, and
chlorOplast extracts from spinach and algae (Tien, 1968a,

b). This pigmented model membrane can show photovoltaic

effect and photoconductivity. Studies on the mechanisms
Of these processes may contribute significantly toward the
understanding of the initial processes of photosynthesis.
For the explanations of the observed photoresponses, Tien
(1968a, b) prOposed a possibility of the existence not
only of ionic but also of electronic charge carriers. It
was further prOposed that if electrons and holes could be
generated by light, the chemical work also could be done
by these charge carriers. Because of a lack of knowledge
of the nature of the charge carriers in the dark state as
well as in the light state, these alternative explanations
still remain undetermined.

In the following year, Hesketh (1969) confirmed
Tien's observations and proposed that similar electronic
charge separation would result in the differential
charging-up of the electric double layers under the effect
of an applied electric field. Tien and Verma (1970)
strengthened the possibility of the existence of the
electronic charge carriers, observing the enhancement of
the photovoltaic effect by placing proper electron acceptors
and/or donors. Indeed, Tien and Verma (1970) pointed out
the similarity between the mode of the photoresponses of
chlorOphyll BLM and those of solar cells.

In an attempt to form photosensitive BLM, inorganic
ions were used by several workers. These include ferric

ions (Tien, l968d; Rosenberg and Pant, 1969) and iodide

ions (Kay and Chan, 1969; Mauzerall and Finkelstein, 1969).
In all of these cases electronic charge transfer processes
were assumed to take place, presumably at lipid/water
interfaces of BLM, similar to the suggestion made by Tien
(1968a, b) in the case of chlorophyll BLM.

If, indeed, BLM are capable of having photogenerated
electronic charge carriers, then the coupling of such an
electronic potential to ionic potential (if any exists) may
exhibit biphasic photopotentials similar to the early
receptor potentials of photoreceptors.

In spite of the recognition of the fact that a
membranous structure plays an essential role in the genera-
tion of early receptor potentials, the temperature depend-
ence of the biphasic mode of the early receptor potentials
in visual receptors has been interpreted only as the
direct manifestation of the different rate of the formation
of rhodOpsin derivatives by the action of light at various
temperatures. Studying the linkage between the photo-
regeneration of rhodOpsin from its photoderivatives to that
of the early receptor potential with elaborated techniques,
Cone (1967) and Cone and Cobbs (1969) proposed that the
fast component of the biphasic responses was spectro-
photometrically related to the transition from rhodOpsin
to metarhodOpsin I, and the slow component to that from
metarhodopsin I to metarhodopsin II. A similar proposi-

tion was presented independently by Pak and Boez (1967).

Although Arden, Bridges, Ikeda, and Siegel (1968), from
thermodynamic arguments, preferred the formation and
decay of prelumirhodOpsin rather than metarhodOpsin I,
an essentially same mechanism was also proposed by Haggins
and McGaughy (1967) for the squid retina. However, this
mechanism is still unsatisfactory since it failed to
explain how such an intramolecular charge rearrangement
can develop not only the membrane potential but also the
potential which can be recorded by the electroretinogram
techniques, as pointed out by Wald (1968). This one-
sidedness of the studies on the early receptor potential
was inevitable, due to the experimental limitation in
using whole eyes and retinas.

The ubiquitous nature of the early receptor potential
of the photoreceptors in the pigmented animal and plant
tissues strongly suggests that such a photoresponse could
be demonstrated by a model membrane system with a proper
biological pigment. One of the best biological membrane
models at the present time is the BLM, especially for the
purpose of measuring electrical responses. The incorpora-
tion of all-trangfretinal into BLM would provide an
excellent membrane model for studying the mechanism of the
generation of the early receptor potential of visual
receptors (Tien, 1968a). A close similarity to the early
receptor potential and not to the early components of the

electroretinogram (the early phases of P-III) of the

photoresponses of all-trans-retinal BLM can be easily

 

recognized, since the membrane model lacks the proteinous
component. Thus, it may not be capable of inducing
selective transport of K+ or Na+ across the membrane. Such
an ionic selectivity can be achieved after a specific
modification of the BLM, as extensively reviewed by Mueller
and Rudin (1969). Furthermore, the absence of phospha-
tidylethanolamine in the present model membrane also
excludes the ionic selectivity as shown by Bonting and
Bangham (1967). If this similarity is demonstrable, the
photoresponse of all-transfretinal BLM will show the
biphasic response whose mode of generation would depend
strongly on temperature, in a similar manner to the early
receptor potential.

The purpose of this work is to explore the possi-

bility of the concept that the early receptor potential

is a membrane phenomenon and is to be understood not

only in terms of the photochemistry of the pigment present
in the membrane but also in terms of the factors which
affect the membrane prOperties. Specifically the following
questions are asked:

1. Is it possible to demonstrate temperature-
dependent biphasic photoresponses with all-
Eransfretinal BLM?

2. What is the mechanism by which temperature can

influence the mode of the photoresponses?

It was successfully demonstrated that all-Eransf
retinal BLM showed temperature-sensitive biphasic photo-
responses with striking resemblance to the early
receptor potential of visual receptors and pigmented
animal and plant tissues. It was also shown that the
mobility of hydrogen ions was the major factor which
influenced the mode of biphasic photoresponses as temper-

ature was varied.

CHAPTER II

LITERATURE REVIEW

Early Receptor Potential

 

Since Brown and Murakami (1964a) discovered the
early receptor potential in the monkey retina, this
potential has gained the enormous interest of visual
physiologists, for this is the earliest electrical
response which can be recorded from the eye. An elucida-
tion of its generation mechanism is believed to contribute
to the understanding of the initial processes of visual
excitation.

Like the intra-retinal recording of the early
receptor potential of Brown and Murakami (1964a) the early
electroretinogram was recorded by Cone (1964) with the
albino rat eye and the frog eye. It was concluded that
the two types Of recording showed the same electrical
response (Cone, 1964; Brown and Murakami, 1964b).

The early receptor potential has the following
features distinct from the electroretinogram. The early
receptor potential is:

1. unaffected by anoxia (Brown and Murakami,

1964a; Cone, 1964),

10

2. less sensitive to the ionic environment
(Pak, 1965; Brindley and Gardner-Medwin, 1965),
and

3. found in the degenerating rat eye, from which

no electroretinogram can be recorded (Arden,
Ikeda, and Siegel, 1966).

The exact relationship of the early receptor poten-
tial to the generation of the electroretinogram is still
under investigation, as well as the mechanism of the
generation of the early receptor potential itself. At
the present time, however, experimental evidence indi-
cates not only that there is a close relationship between
two types of photoresponses but also that the early
receptor potential could result in the generation of the
first wave of the electroretinogram (P-III). For
example, Cone (1964) observed that the early receptor
potentials had:

1. the same spectral sensitivity as the b-wave of

the electroretinogram,

2. its amplitude linearly proportional to the

rhodOpsin bleached by flash, and

3. been found in the eye which could not generate

the electroretinogram as also shown by Brown
and Murakami (1964a) and later by Arden,

Ikeda, and Siegel (1966).

ll

Particularly, Gage, Crawford, and Brown (1968) observed
that while the early receptor potential was independent of
ionic changes in the aqueous medium, the early phase of
the electroretinogram was clearly ion dependent. From this
Observation they suggested that the transition from the
first two non-ionic potentials to the ionic potential of
the third phase would be the fundamental process underlying
the excitation of photoreceptor cells.

The early receptor potential was also found in the
human eye by Yonemura and Kawasaki (1967) and Galloway (1967).
The clinical application of the early receptor potential had
the following advantages over many other tests of visual
functions: the independence from the neural events would
allow a differential diagnosis for pathological conditions
affecting the peripheral structures of the visual system and
would give direct information on the visual pigment in the
rods and cones. And the dark adaptation would not be
required.

The Effect of Temperature on the
EarIy Receptor PotentIaI

 

 

Shortly after its discovery by Brown and Murakami
(1964a) one of the remarkable properties of the early recep-
tor potential was reported. At room temperature the early
receptor potential had two components in the frog eye
(Cone, 1964) and in the monkey retina (Brown and Murakami,
1964b). The first corneo-positive phase was observed with-

out detectable latency, had a peak time of about 100 us,

12

and was followed by the second corneo-negative phase with
a peak time of about 900 us. Since the sign of two com-
ponents depends on the position of the recording electrodes,
Cone (1965) later proposed the use of R1 and R2 (R:
Receptor Potential) for the first and second components of
the early receptor potential, respectively. Pak (1968)
used R1 and R2 for the two components of "rhodOpsin
responses." It is due to this discovery of the biphasic
nature of the early receptor potential that the under-
standing of the mode of its generation has been strengthened,
since various experimental conditions are now known to
influence R1 or R2 independently.

Temperature is the most effective and first dis-
covered parameter in controlling the mode of generation
of the early receptor potential. Pak and Cone (1964)
demonstrated the different temperature dependence of R1
and R2 in the albino rat eye, and discovered that lowering
the temperature sufficiently abolished R2 and isolated R1,
which could be detected even at a subzero temperature of
-35°C (Pak and Cone, 1964; Pak, 1965; Pak and Ebrey, 1965).
Similar results were observed in the excised goldfish eye
(Pak and Cone, 1965), and in the frog eye (Brindley and
Gardner-Medwin, 1966).

The occurrence of temperature-sensitive biphasic
photoresponse is not restricted to the retina. It is now

established that the early receptor potential can be found

13

in non-ocular pigment epithelium and the leaves of green
plants (Brown, 1965; Brown, Watanabe, and Murakami, 1965;
Brown and Gage, 1966a, b; Arden, Bridge, Ikeda, and
Shiegel, 1966b; Arden, Ikeda, and Shiegel, 1966a, b;
Becker and Cone, 1966; Ebrey, 1967, 1968). In spite of
the minor differences in response time and sensitivity
toward temperature variations, all photoresponses observed
in these pigmented tissues show essentially the same
features of the early receptor potential and, hence,
indicate the common mechanisms of the origin of the
initial photoresponse in all pigmented biological tissues.
The Necessity of the Membranous

Structure for the Generation of
the Early Receptor Potential

 

 

 

At the present time the early receptor potential
has not been obtained with aqueous suspension of rhodOpsin
or any other pigments or even with the aqueous suspension
of rods and cones. Instead, the Optimum early receptor
potential can only be observed in membranous structures
with high electric resistance and proper orientation of
the pigment molecules.

The requirement of the organized pigment arrangement
in membrane structure for the generation of the early
receptor potential was indicated by its abolishment after
freezing the squid retina to -80°C and rewarming it
(Hagins and McGaughy,l967), or after heating a vertebrate

eye to 58°C and cooling it (Cone and Brown, 1967). Under

14

these conditions it was found that, although the absorption
spectrum of rhodOpsin was not affected, the microspectro-
photometric measurement indicated a complete disorienta-
tion of rhodOpsin molecules from the intact membrane, as

if rhodOpsin molecules were in a solution (Cone and

Brown, 1967).

The high efficiency of cone early receptor potential
was explained by Goldstein (1967, 1968, 1969) and Gold~
stein and Berson (1969) as an indication of the structural
difference between cones and rods. In rods the presence
of outer membrane separated from pigmented discs was
rather inefficient for generation of the early receptor
potential. From this observation Goldstein concluded
that the membranous structure was necessary and that the
early receptor potential was considered to be generated
across the pigmented membranes.

The involvement of the cell membrane in generation
of the early receptor potential was also indicated by the
observation made by Brown and Crawford (1966, 1967a, b)
with the pigment epithelium. After penetrating a pigment
epithelium cell, the membrane potential declined because
of possible membrane damage by the microelectrode, and R2
of the responses decreased more rapidly than R1. This
could suggest that the second phase was specifically

dependent upon the integrity of the cell membranes.

15

The Necessity of the Mediator for
the Generation of the Early
Receptor Potential

 

 

 

The great influence of the front membrane on the
mode of generation of the negative phase of the early
receptor potential of frog pigment epithelium led Crawford,
Gage, and Brown (1967) to the conclusion that the genera-
tion of the early receptor potential occurred across the
front membranes. They felt that the involvement of the
membranous structure was a secondary process following
the primary response that had to take place in the melanin
granules, which were not the membrane component and were
not continuous to the cellular membrane. From this they
asked the question of how the primary event could be
transferred to the membrane and initiate the R2 phase.
Furthermore, they suggested the generality of this type of
energy transfer from the pigment granules to the membranous
structure in all varieties of photoreceptors.

It is interesting to note that in the retina which
contains both rods and cones (e.g., frog and monkeys) the
early receptor potential has dominant cone origin
(Goldstein, 1967, 1968, 1969; Goldstein and Berson, 1969).
Because of the fact that cone pigments consist of only 10%
of the total pigment content in these eyes, Goldstein
(1968) concluded that the dominance of the cone origin in
the early receptor potential could be due to the structural

difference between cones and rods. The outer membrane of

l6

rods could mask the early receptor potential, which other-
wise could be detected like that of cones. On the other
hand, the eyes with pure rods generate an equally strong
early receptor potential as those with pure cones. Indeed,
Pak and Ebrey (1966) carefully studied the dependence of
the early receptor potential of animals with pure rod
retina and pure cone retina as a function of temperature,
stimulus intensity, and spectrum, and concluded that there
exists an essential similarity between the pure rod retina
and the pure cone retina. According to Cohen (1968), the
disc membranes are enclosed and separated from the outer
membrane of the frog rods. This observation is in support
of Goldstein's (1967) proposition; however, it cannot
explain that of Pak and Ebrey (1966). The outer membrane
of the rods of the pure rod retina may have a somewhat
different structure from that of the vertebrate rod
structure (Dowling, 1967); however, this View is unlikely.
Pedler and Tilley (1967) observed incomplete membrane con-

tinuity between the disc and outer membranes of the rods

 

of the guinea pig (Cavia), the baboon (Papio Cynocephalus),

the toad (Xenopus laevis), and the albino rat (Rattus, v.

 

albino).

Apart from the above electronmicroscopic observations,
there is some evidence on the nature of the outer membrane
of the rod outer segments. According to Adams and Hagins

(1960), the inter-disc space is ionically continuous to

17

the outer space of the frog rod outer segments and seems
to be electrically transparent. Dewey, Davis, Blasie, and
Barr (1969) observed the existence of a diffusion pathway
into the frog rod outer segment of sufficient size to
allow molecular species as large as antibody molecules,
antirhodOpsin serum and fluorescein-labeled sheep anti-
rabbit y—globulin to pass through. If these observations
can be taken as indications of the inertness Of the

outer membrane of the rods, the disc membranes of the
rods could function in exactly the same manner as the
disc membranes of the cones, which are continuous to the
outer membranes (Dowling, 1967).

As described above the outer membrane of the rod
outer segment does not seem to play an important role in
the generation of the early receptor potential. Further-
more, the intracellular recording closely resembles the
extracellular early receptor potential. These observa-
tions indicate that it is not likely that asymmetric
orientation of the rhodOpsin molecules relative to the
surfaces of the retina alone leads to the generation of
the early receptor potential as suggested by Brindley and
Gardner-Medwin (1966) and by Lettvin (1965). Rather,
these Observations indicate that the orientation of the
photOpigment relative to the photoreceptor membranes is
an important factor in determining the direction and

magnitude of the early receptor potential, as suggested

18

by Smith and Brown (1966) and Hagins and McGaughy (1967).
Thus, it is very likely that in rods and cones the disc
membrane itself is capable of generating both positive
and negative components of the early receptor potential,
unlike the pigment epithelium. This is in accord with
the common knolwedge that the disc membranes contain most
of the visual pigments in the rods and cones (Dewey,
Davis, Blasie, and Barr, 1969).

There seem to be differences in the functions of
the membranous structures in generation of the early
receptor potential in various types of photoreceptors as
well as the structural differences shown in Figure 1.

In the pigment epithelium and the leaves, the cellular
membrane (which is referred to as the membranous structure
by the investigators) is not primarily pigmented and the
initial photoevent has to be transferred to the cell
membrane from the discontinuous pigment granules. On

the other hand, in the rods and cones the disc membrane
(which is also referred to as the membranous structure by
the investigators) contains the pigment molecules without
any auxiliary structures. The consequence of this
difference between two types of photoreceptors on the
function of the membranous structure in the generation of
the early receptor potential can be seen in the following

two points.

19

Figure l.--The schematic diagrams of the photoreceptor
cells from which the early receptor potential
has been recorded. Cell membrane, across
which the early receptor potential develops,
does not always contain pigments. In the outer
segment of cones and rods (a) the pigment is
incorporated in the cellular membrane; in the
pigment epithelium cells of eyes and skins (b)
the pigmented granules (g) are isolated from
the cell membrane, and in the plant cell (c)
the pigment is incorporated in the membranous
structure (t) in the chloroplast (e), which is

isolated from the cell membrane.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

21

The first point is that, although the understanding
of the mechanism of generation of the early receptor
potential of the pigment epithelium and the plant cells
would certainly contribute to the understanding of that
of the rods and cones, the above characteristic difference
has to be taken into consideration. Furthermore, this
difference could be so important that the substitution of
the former photoreceptors for the study of the latter
photoreceptors may not be desirable. This point has to
be seen in future works with various photoreceptors.

The second point is the necessity of the mediator
of the early receptor potential generation, which not
only can develop the cross-membrane potential but also
can interact or be transferred to the membranous structure
located away from the original site of the formation of
the mediator. The nature of the mediator is totally
unknown at the present time. Nevertheless, the identifi-
cation of such a mediator may help us to understand the
mechanism of the generation of the early receptor
potential in all photoreceptors.

The Selection of the Membrane
ModeITSystem

 

 

The Importance of Membrane
Models in the Studies on the
Early Receptor Potential

 

 

 

In order to understand generation of the early

receptor potential it is necessary to investigate changes

22

in the membrane properties upon illumination. However,
it is almost impossible to carry out intracellular
recording of the rods and cones. Brown and Wiesel (1961)
feel that the rod outer segment is so small (5-6 u
diameter and 50 u length) that rods are relatively free
to move when contacted by an electrode, due to the
hardened structure of multilamellar formation. Further-
more, there is the difficult problem of placing an
electrode tip in an intrasaccular width of less than

100 A. Thus, there is no way of being sure of making an
"intracellular" recording in the outer segments of rods
and cones from any direction with respect to the orienta-
tion of the outer segments.

Smith and Brown (1966) reported the intracellular
recording of the early receptor potential from Limulus
eyes. Even with such a simple eye the complexity of the
organization of membranous structure exists. Whether the
microvilli are connected to retinular cells is still
controversial (Walther, 1966). Assuming the continuity
of the cellular membrane from the retinular cell to that
of the microvilli as reported by Eguchi and Waterman
(1966), the extensive localization of visual pigment on
microvilli (Langer and Thorell, 1966) suggests that the
intracellular recording from the retinular cells is
equivalent to the intracellular recording with an electrode

in the inner segment of the rods and cones of vertebrate

23

eyes. More recently, Lasansky and Fuortes (1969) recorded
the intraretinular cell and intravacuole electric responses
with the leech eye, and proposed that the site of the
origin of the electrical response in the visual cells

would be the region of the microvilli. Again, since the
diameter of the microvilli is less than 1000 A (Eguchi

and Waterman, 1966), intramicrovillus recording is
extremely difficult.

As shown above, the real photoreceptor systems
present technical problems for the investigation of the
function of pigmented membranes in generation of the early
receptor potential. The other pigmented cells such as
epithelium and plant cells contain pigment granules or
chloroplasts, but the cell membrane itself does not con-
tain the pigments. The mode of the early receptor poten-
tial generation in these cells, therefore, has to be
different from that of rods and cones in some aspects.
Thus, the model membrane system with visual pigments is
invaluable for the investigation of the mode of generation
of photoresponses, being sure of a contribution from only
the pigments incorporated in the membranous structure.

In the following three sections the structure of the
disc membranes, which should be reproduced by ideal
membrane models, and the works on photosensitive lipid
monolayers and bilayers will be presented in order to help
select an ideal membrane model for the study on the early

receptor potential.

24

The Structure of the Membranes
of Rods and Cones

 

 

From low-angle x-ray diffraction studies Blasie and
Worthington (1969) concluded that the pigment molecules
of retinal receptor disc membranes would have a core of
uniform electron density 38-40 A in diameter with a total
diameter of 44-46 A, and would occur in a plane or two
dimensional liquid-like arrangement within the disc
membranes. Blaurock and Wilkins (1969) specified the
position of the pigment on the outer surface of the
bilayer lipid membranes in a symmetric manner. Further,
they proposed that the thickness of the pigment molecules
could be less than 10 A, still occupying the area per
particle found by Blasie and Worthington (1969). Accord-
ing to Blaurock and Wilkins (1969), the unhydrated sub-
stance in a membrane was about 55 A thick, assuming a
uniformity in the membrane, and lipid occupied 26 A of
which 18 A was estimated to be hydrocarbon chains.

The low angle x-ray diffraction data of biological
membranes provided the thickness of hydrocarbon layers
in a range of 20-30 A (Finean, 1969). The minimum thick-
ness of the hydrocarbon layers would exist due to the
ionic interaction of the polar head groups of phospho-
lipids, and was estimated as 15 A for phosphatidylethanola-
mine molecules and phosphatidylcholine molecules (Finean
and Millington, 1955). In this liquid crystalline state,

phosphatidylethanolamine had a layer thickness of about

25

35 A and the pure hydrocarbon region would be only 20 A
thick. Thus, it may be concluded that the retinal disc
membrane has a similar basic membrane structure to the
bilayer lipid membrane coated with pigment protein
molecules.

It is still uncertain how retinal is arranged in
the disc membrane. Since rhodOpsin has a hydrophobic
core and several binding sites between its opsin part
and the retinal chromophore part (Abrahamson and Ostroy,
1967) and since retinal itself is lipid soluble, it is
likely that the retinal moiety of rhodOpsin molecules is
located in the lipid-rich layer of the disc membrane as
suggested by Maeda and Isemura (1967). Wolken (1961,
1966), on the other hand, located the retinal molecules
at the hydrophilic portion, or the outer layer of the

membrane.

Photosensitive Monolayers

 

Hyono, Kuriyama, Tsuji, and Hosoya (1962) spread
rhodOpsin extracts with dilute ethanol or digitonin
solutions at air/water interfaces. The aqueous phase was
a phosphate buffer solution of pH 6.8. Although the
rhodOpsin solution extracted with a dilute ethanol solu-
tion gave a suitable monolayer film, both rhodOpsin
monolayer films gave an increase in surface pressure upon
illumination with a light flash. After the bleaching of

rhodOpsin, this increase in surface pressure upon

26

illumination completely disappeared. Azuma and Takagi
(1966) pointed out that under appropriate experimental
conditions illumination had no effect on the surface
pressure of the rhodOpsin monolayer at air/water interfaces.

The surface active property and the molecular
orientation of vitamin A derivatives at air/water inter-
faces have been well established (Bangham, Dingle, and
Lucy, 1964; Anderson, Roels, Dreher, and Schulman, 1967;
Roels and Shah, 1969). The conjugated chains are immersed
in air or a residual monolayer of hydrophobic solvent,
and the aldehyde, hydroxyl, and carboxyl groups are
directed toward the aqueous phase. The similar molecular
orientation in bilayer lipid membrane was also demon-
strated to exist by the studies on lecithin myelinics
(Dreher, Schulman, Anderson, and Roels, 1967; Anderson,
Roels, Dreher, and Schulman, 1967; Roels and Shah, 1969).

The action of light on the monlayer of retinal was
investigated by Maeda and Isemura (1967). All-ppaps-
retinal and ll-gig-retinal showed changes in the surface
pressures required to keep the film area constant after
illumination. The maximum change was Observed after
illumination for 30 seconds. The all-ppaps—retinal film
showed an increase in the surface pressure while the 11-
gipfretinal showed a decrease. They interpreted an
increase in the case of all-Epapg-retinal as a film

expansion by the isomerization to the most hindered

27

conformation, since the ll—cis isomer showed a larger
limiting area and lower collapse pressure than the all-

trans isomer.

 

Beta~carotene films at air/water interfaces were
prepared by spreading a hexane solution of the pigment
over 10-3 M phosphate buffer solution at pH 8 (Colmano
and McGlone, 1961). Colmano (1962) was successful in
obtaining the absorption spectrum of the monolayers of the
beta—carotene mixed with chlorophyll a and chlorophyll b.

Monolayer studies have been carried out at air/
water interfaces and, hence, have been limited to the
elucidation of the mechanical properties of the mono-
layers with rhodOpsin and retinals. Investigation of the
electric properties of photosensitive films had to wait for
the successful formation of BLM.

Photosensitive Bimolecular
Lipid Membrane

 

 

The formation of photosensitive BLM with visual
pigments was tried by several workers. Mueller, Rudin,
Tien, and Wescott (1963) observed that the BLM formed
from chloroform-methanol solutions of brain lipids or
proteolipids became electrically excitable after the
adsorption of water-soluble molecules extracted from
bacterially fermented retina. Takagi, Azuma, and
Kishimoto (1965) formed bimolecular lipid membranes with

the digitonin-free rhodOpsin lipo-protein extracted from

28

cattle retina. The resistance of the membrane was 103 —
105 ohm cm2 with l M KCl solutions in both of the chambers,
and increased to 106 ohm cm2 when one aqueous phase was

replaced with isotonic CaCl Excitability, however, was

2.
not observed with this preparation. Leslie and Chapman
(1967) incorporated beta-carotene, retinal, or retinol
into phospholipid BLM and observed that thinning time
was affected due to this additional pigment component.
The existence of beta-carotene in the membrane was
indicated by a specially designed absorption spectroscopy.
But, again no photoexcitability was observed with their
BLM containing beta-carotene, retinal, or retinol.

The initial account of the present investigation
was reported elsewhere (Tien and Kobamoto, 1969). Photo-
voltaic effect and photoconductivity were Observed by
incorporating all-trans-retinal, 9-pis-retinal, all-trans-
retinol, 9—gigfretinol, and beta—carotene in 10-1 M KCl
solution. Pant and Rosenberg (1970) reported photo-
conductivity effects in beta-carotene, retinal, and

retinols contained in BLM in the presence of iodine-

iodide.

CHAPTER III
THE CONSTRUCTION OF THE MODEL MEMBRANE

SYSTEM USED IN THE PRESENT WORK

All-Trans-Retinal Bimolecular
Lipid Membrane

 

 

An ideal model BLM would be formed with the lipids and
pigment which are known to be the constituents of the visual
receptors. Lipids in the outer segment of the rods con-
stitute 20% - 40% of the dry weight, most of which is
phospholipid (Wolken, 1962). Cholesterol exists in the
disc membrane of the outer segment but only in trace amounts
(Feldman, 1967). The membrane forming solution used in the
present work was prepared by dissolving lecithin and oxidized
cholesterol in the solvent pfoctane and p—dodecane. The use
of pure hydrocarbon was inevitable and the trace amount of
solvent residue was assumed to remain in the final BLM as
in a phospholipid BLM (Henn and Thompson, 1968). Neverthe-
less, the major lipid constituents of the present membrane
are the surface active molecules; lecithin and oxidized
cholesterol. Phosphatidylethanolamine was not used in the
present membrane forming solution since undesirable ionic
selectivity would be induced by the formation of all-Egapgf
retinal—phosphatidylethanolamine complexes while lecithin
was known to be free from such a complication (Bonting and

Bangham, 1967).
29

3O

All-trans-retinal, instead of rhodOpsin or ll-cis-

 

retinal, was selected as a visual pigment in the present
work because the pure commercial product was available, it
was the most stable conformer of the retinals, and it lacked
the regenerative and degenerative factors found in rhodOpsin,
which would complicate the mode of generation of the photo-
responses. The use of such a simple pigment does not lessen
the potential significance of the present model membrane,
since the mode of generation of the photoresponse of the
type with which the present work is concerned is not depen-
dent On the kinds of pigment investigated, taking into
account variations in spectral sensitivity.

In animal and plant tissues only pigmented areas yield
responses; for instance, melanin containing skin, or pig—
mented (non-albino) iris or pigmented epithelium. In these
tissues melanin is the light absorbing compound. Likewise,
parts of variegated leaves which do not contain chloroplasts
do not yield responses (Arden, Bridges, Ikeda, and Siegel,
1966b). Thus, it is clear that the mode of generation of
the biphasic responses does not depend on the types of pig-
ments. From this fact it can be concluded that the gig-

trans isomerization of retinal itself is not responsible for

 

the generation of the early receptor potential unless the
isomerization induces the secondary effects which happen to
be the common physicochemical effects to all the pigments.

It is also clear that the formation of the stepwise rhodOpsin

derivatives is not likely to be the main mechanism of the

31

generation of the early receptor potential, since such a
distinct sequence of photoproducts is not known in melanin
and chlorophyll. Thus, a clue to the elucidation of the
mechanism of the generation of the early receptor potential
has to be present in the early processes of light action
common to all photosensitive biological compounds found in
membranous structures.

The chemical structure of melanin is still unknown
(Hempel, 1966; Mason, 1967), but it is well established that
in vivo melanin is strongly associated with proteins (Seiji,
Kishimoto, Birbeck, and Fithatrick, 1963; Fitzpatrick,
Quevedo, Levene, McGovern, Mishima, and Oettle, 1966).

Since chlorOphyll and retinal also exist in associated form
with proteins, one may ask whether protein component can
play any significant role in the generation of the early
receptor potential. At the present there is no evidence
that the absorption of photons by these chromOproteins leads
to the dissociation of the chromophores from any protein
components except rhodOpsin. Cone (1967) observed that a
completely bleached retina with dissociated retinal or
retinol and opsin could not generate the early receptor
potential. On the other hand, Crawford, Gage, and Brown
(1967) demonstrated that, after soaking in a solution of
synthetic melanin (protein free), filter papers could show
the photoresponses similar to the early receptor potential
of the frog pigment epithelium. Glutaraldehyde fixation of

the retina removes the R2 component selectively (Arden,

32

Bridges, Ikeda, and Siegel, 1968). This evidence could also
support the importance of the protein component in the gen-
eration of the early receptor potential. However, it is also
possible that the membrane structure as a whole could be
affected with such a strong protein denaturing agent. The
decay of the magnitude of the early receptor potential by
extreme illumination also cannot give conclusive evidence
since it occurs in all three photosensitive systems regard-
less of whether the dissociation of the components occurs

or not. Thus, at the present time it is not likely that the
photodissociation of retinal from opsin is the primary cause
of the loss of the early receptor potential generation of
the retina. Rather it is more likely that the temporal
detachment of retinal from the disc membrane or the random
orientation of the retinal molecules in the disk membrane
after the bleaching may be the primary cause of the loss of
the early receptor potential in the completely bleached
retina. This interpretation is in accord with the necessity
for organized pigment arrangement in the membranous struc-
ture. Furthermore, the nature of the association between
the chromophore and the protein is not known in the three
pigment systems. Thus, it is more fruitful at the present
stage of investigation to examine whether the pigment itself
can generate the photoresponse similar to the early receptor
potential if the pigment can be incorporated into a mem—
branous structure. It is a matter of interest to note that

all three pigments (retinal, chlorophyll, and melanin) have

33

at least one C=O group, which can act as a hydrophilic group
when present at a lipid/water interface.

Some of the basic prOperties of the all-Egapg-retinal
BLM in symmetric 0.1 M KCl solutions together with other
carotenoid BLM were reported previously (Tien and Kobamoto,
1969). The important feature to be noted is that the
electrical properties and the mode of photoresponses are
quite similar regardless of the kinds of pigment incorpo-
rated, with the exception of beta-carotene. It has been
emphasized that not only the presence of photosensitive
molecules in the lipid phase, but also the interaction of
the molecules with the aqueous phases seemed to play an
important role in the generation of the photoresponses

(Tien and Kobamoto, 1969).

Aqueous Phases

 

In the absence of electrical and chemical asymmetry
across the BLM it was found at an earlier stage of this
work that the photoresponses were small and not very
reproducible. This is probably due to the variation of
microenvironment around the membrane or around the
Plateau-Gibbs border which supports the membrane.
Asymmetric charge distribution may originate from the
uneven distribution of lipid at the border or slight
hydrostatic pressure gradient across the membrane, which
can be easily generated by the removal of a drOplet of

water by the lipid solution injector. In order to remove

34

this difficulty a proper substance for the generation of
membrane asymmetry was looked for. FeCl3 was selected
since it gave the significant enhancement of the photo-
response as well as reproducible resting potentials.
Furthermore, in spite of some controversy over the
absence of any trace metal in rhodOpsin (Fukami, Vallee,
and Wald, 1959) or its presence (Fujishita, 1962), it is
almost certain that the outer segment of rods contains
some metal: McConnell (1965) observed non-enzymic
oxidation of ascorbate and attributed it to the presence
of some metal associated with the disc of bovine retinal
outer segments. Since Fujishita (1962) identified ferric
ions in digitonin-rhodOpsin complex, the use of FeCl3

as an agent for making chemical electrical asymmetry not
only may be justified but also may elucidate the role of
the trace metal in visual excitation. The action spectrum

of the FeCl system is exactly the same as that of the

3
non-FeCl3 system (Tien and Verma, 1970), implying that

FeCl functions as a sensitizer for the development of

3

the photoresponse of all—trans-retinal BLM.

CHAPTER IV

EXPERIMENTAL

Lipid Solutions

 

The oxidized cholesterol solution used in the
present work was prepared from freshly recrystallized
cholesterol obtained from Eastman, Rochester, New York
and p-octane (pure grade, 99% minimum) from Phillips
Petroleum Co., Bartlesville, Oklahoma by the method of
Tien, Carbone and Dawidowicz (1966). In this method 12 gm
of the recrystallized cholesterol was added to 300 ml of
p—octane in a l—L flask and heated to the boiling point
(125°C) with molecular oxygen bubbled at a rate of 100-
125 cc per minute for 5.5 to 6.0 hours. Then, the
mixture was transferred to a flask and kept overnight at
5°C. The final solution kept at room temperature was
colorless with white precipitation.

A lecithin solution was prepared by dissolving 1 gm
of L-alpha-lecithin (chromatographically pure) from egg
(General Biochemicals, Chagrin Falls, Ohio) into 100 ml
of p—dodecane (pure grade, 99% minimum) from Phillips

Petroleum Co., 110 m1 of oxidized cholesterol solution,

35

36

and 90 ml of 1% cholesterol in pfoctane solution. The
stock solution was kept in the freezing compartment of a

refrigerator.

Membrane Forming Solution

 

The all-trans-retinal solution prepared with
oxidized cholesterol solution formed a BLM with a life-
time of not more than 10 minutes. With lecithin solution
the all—trans-retinal BLM gave no apparent stable state
although it had a lifetime longer than 30 minutes. The
most satisfactory membrane forming solution used in the
present work was prepared by dissolving 0.025 gm of all-
trans-retinal to 0.25 ml of a one-to-one mixture of the
oxidized cholesterol solution and the lecithin solution.
The membrane forming solution was freshly prepared each
day since the concentration of all-trans-retinal was
close to the saturation concentration, and the evaporation
of hydrocarbon solvent Often resulted in the precipitation
of the retinal. The membrane stability was recuded if

the solution had been kept over night.

Chemicals

 

All chemicals used for making aqueous solutions
were of reagent grade as obtained from the manufacturers.
Double distilled water from an all-glass still was used

throughout the present work.

37

Aqueous Solutions

 

In all cases the hole for the membrane formation
was filled with a drOp of lipid solution after placing
an aqueous solution into the inside and outside chambers.
In order to create the concentration gradient of identical
ions or different ionic species at the same or different
concentrations across the membrane, 0.05 ml or 0.1 m1 of
the original solution was removed with a micropipette and
the same amount of a proper solution was added each time

to insure the hydrostatic equality across the membrane.

Membrane Formation

 

The membrane forming cell used in this work is
similar to that of Mueller, Rudin, Tien, and Wescott
(1962a, b). Shown in Figure 2 the outer chamber is a
glass cup and the inner chamber is a Teflon cup (Hanai,
Haydon, and Taylor, 1964; Tien and Diana, 1967), which
separates the inner and the outer chambers except for a
small hole for the membrane formation (Figure 2). The
volume of aqueous solution put in the chambers in most
experiments was 10 ml for the inner chamber and 40 ml
for the outer chamber. The diameter of the membrane-
forming aperture of the Teflon cup was 0.17 cm. Due to
this small diameter of the hole the resulting membrane
was minute. The membrane formation was, therefore,

observed with a low power microscope under dim red light.

38

Figure 2.--The apparatus for the formation of BLM and
the measurement of membrane potentials. The
inner chamber was made from a Teflon cup
(10 ml, T). A portion of the wall of the cup
was machined down to about 0.0025 cm and a
hole (0.17 cm, h) was punched through it.

The outer chamber was made of glass (g),

the front portion of which was flattened to
facilitate the observation of the membrane.
The bottom chamber (C) was filled with water
for temperature control. For the measurement
of potentials a pair of calomel electrodes
(e) were connected to an electrometer (E) and

a recorder (R).

40

To form BLM, both chambers were first filled with
aqueous solution, and 0.01 ml of all-Epaps—retinal
solution was then injected into the hole with a micro-
meter syringe. The use of the micrometer syringe for
membrane formation was first experimented with by Tien
and Diana (1967). The best performance of the syringe
was achieved when the tip of the fine capillary attached
to the syringe was slightly smaller than the diameter
of the membrane forming hole. The ease of the formation
and membrane stability were also achieved by wetting the
hole and the surrounding area with the lipid solution

prior to the membrane formation.

Potential Measurements

 

The apparatus and procedures used in this study for
the measurements of membrane potential and resistance
were described by Mueller, Rudin, Tien, and Wescott
(1962a, b) and Tien and Diana (1967). The membrane

potential was studied in a set up as illustrated in

Figure 2. The cell arrangement may be represented as
follows:

Saturated Aqueous Membrane Aqueous Saturated
calomel solution solution calomel
electrode (inside) (outside) electrode

A potential across the membrane was measured with an

electrometer of high input impedance (Keithley, Model 6108)

41

through a pair of calomel electrodes via saturated KCl
salt bridges. The output of the electrometer was fed
into a strip chart recorder (Servowriter II, Texas

Instruments, Inc., Houston, Texas).

Resistance Measurements

 

A d.c. membrane resistance was obtained by applying
a potential between the electrodes and introducing a
known resistance in series with the membrane determined
(Figure 3). Polarizing potentials were applied to the
membrane using the instrument purchased from Electronics
for Life Sciences, Rockville, Maryland. With the use of
this instrument, the input resistance could be varied
from 105 to 109 ohms and polarizing voltage from 0 to
5.4 V.

The membrane resistance is calculated according to

Ohm's law for the circuit shown in Figure 3, and is given

by:

Rm (EIn X Ri)/(Ei - Em) (1)

where Ri is the input resistance in series with the
membrane, Ei is the applied voltage (usually 20 mV), and
Em is the voltage in the presence of the membrane. For
the best results Ri should be adjusted so that Em/Ei lies

between 0.1 and 0.8.

42

Figure 3.--Circuit diagram for the measurements of

resistance.
Em — electrometer
Rm - membrane resistance
Cm - membrane capacitance
Ri - input resistors (105 - 109 ohm)
E - voltage divider and switch for

applying polarizing potentials

R - recorder

44

pH Measurements and Control

 

pH of the solutions was measured with a Beckman pH
meter and with Beckman standard pH buffers. The adjust-
ment of pH was done with HCl and KOH and an acetate
buffer system when the control of buffer capacity was

required.

Temperature Measurements and Control

The temperature of the aqueous phases was measured
with a thermometer (Will Scientific, Inc., Rochester, New
York) with 0.l°C divisions and a range from -l°C to 50°C.
The thermometers were placed in both of the chambers and
checked repeatedly throughout the experiments. The
temperature of the chambers was controlled by running
water thermostatically regulated by the use of a Haake
circulator (Type Re, Berlin, West Germany), through the
bottom chamber of the cell assembly shown in Figure 2.
The temperature was maintained within 10.05°C. All
experiments were carried out at equilibrium in tempera-

tures.

Light Sources

 

The photoexcitation of the membrane system was
achieved by illumination with a Sylvania tungsten halogen
lamp (DWY, 650 W, Sylvania Electric Products, Inc., Salem,
Massachusetts) in a Sylvania "sun-gun" (Model SGSST). An
Alphax shutter (Wollensak, Rochester, New York) was used

for controlling the duration of the illumination.

45

Filters

The light beam emerging from the aperture of the
lamp housing passed through a heat absorbing filter
(water bath of 10 cm light path), a cupric sulfate
solution filter (2 cm), two colored glass filters, and a
lens before reaching the chamber of membrane formation.
The concentration of cupric sulfate was adjusted to
achieve the effective elimination of infra-red radiation
as demonstrated by Kasha (1948). For the control of
light intensity, gray filters (Carl Zeiss, Inc., New York)
were used. Except for the intensity dependence experi-
ments, photoresponses were induced without using any gray

filters.

Artifacts

 

The illumination of lecithin-oxidized cholesterol
solution did not show any significant photoresponses. The
illumination of one calomel electrode while keeping the
other out of a light beam also did not show any photo-
responses in either the presence or absence of non-
pigmented membrane. Indeed, the illumination of one
electrode even in the presence of all—trans-retinal BLM
did not show the photoreSponses either. Thus, it was
concluded that the photoresponses recorded with the set-up
are true photoresponses of all-trans-retinal BLM, free
from the artifacts of electrodes and the lipid component
itself. Similarly, no photoresponse was observed with the
pigment-free BLM in the presence of FeCl3 in one aqueous

phase.

46

The possibility that heating of water may cause the
charge density gradient across all—ppapgfretinal BLM was
excluded by the observation that there was not significant
change in the temperature of the aqueous phases during
illumination. After a 60 second illumination at room

temperature the temperature changes were less than 0.05°C.

Transference Numbers

 

The membrane potential, Em, due to a differential
migration of ions through the membrane as a result of salt
concentration gradient across the membrane was measured
with the same circuit as shown in Figure 2.

The transference number of an ionic species is
defined as the fraction of the total current carried by
the species. Thus, if t+ and t- are the cation and anion
transference numbers respectively, and i+ and i- the
corresponding ionic currents, then, t+ = i+/(i+ + i-),

t' = i-/(i+ + 1‘), and t+ + t’ = 1.
Assuming the constancy of the transference number

across the membrane, the membrane potential can be

expressed by (MacInnes, 1961):

 

1 i i
a a a
_ RT + M _ _ + Cl _ RT + _ MCI
Em — F [t lnao (l t )ln;6—] — F—(Zt l)1n o
M c1 aMCl

(2)

47

where a is the cationic activity, is the mean ionic

M
activity of the salt MCl, a

aMCI

C1 is the chloride activity, P
is the Faraday constant, R is the gas constant, and T is
the absolute temperature. The superscripts o and i refer
to the aqueous solutions on the outer and inner sides of
the membrane, respectively. The ion activities are
assumed to be equal to the mean ionic activity of the

salt: = a

ac1 M = aMCl'
In each of the transference number experiments, only
solutions of the chlorides of monovalent cations were

used.

CHAPTER V

RESULTS

Dark Membrane Potentials

 

In order to understand the mechanism of the develop-
ment of the photoinduced potentials of all-trans-retinal
BLM, it is necessary to elucidate the mechanism of genera-
tion of the dark or resting membrane potentials. For this
purpose the nature of the ionic charge carriers was
examined by evaluating the transference numbers of the
ions used in the present work.

Shown in Figure 4 are changes in the dark membrane
potentials due to an increase in the salt concentration
in the inner chamber, which was originally set equal to
that of the outer chamber (10_3 M solutions). HCl, KCl,
and NaCl developed negative potentials. The existence of
the linearity clearly indicates the applicability of the
Nernst equation. With the slope of 58 mV per unit pH, the
protonic diffusion potential was far greater than that
of other cations and anions used in the present work.

The transference numbers calculated from Figure 4 are
shown in Table 1.
In the present work ferric ions were used frequently

as electron acceptors. Due to the strong acidic nature of

48

49

Figure 4.--Cation induced potentials as a function of
concentration. The outside concentration was
10-3 M for all salts and the inside concentra-
tion was increased with the addition of l M

solutions.

a - HCl

b - KCl

c — NaCl

Ci - concentration in the inner chamber
C - concentration in the outer chamber

 

I20

100 "

_
O
6

“>93 _o:c0.om a. DO

40'-

20*-

2.0

 

 

 

 

51

TFZXBLE l.--Cationic transference numbers of all-trans-
retinal BLM.

 

 

Change in Equilibrium Transference
iIc>n Potential per 10-fold Number
Concentration Gradient
(mV)
H+ '58.?) 1.0
K+ - 3.3 0.53
Na+ - 1.3 0.51

The values for the potentials were obtained from
‘tlme slope of the plots of Em against loglOCi/CO shown in
I?jxgure 4. The experiment was carried out at room
temperature (22°C - 24°C).

52

the agent, the addition of FeCl3 resulted in the genera-
tion of the dark membrane potential. The magnitude of
this dark membrane potential was identical to those
developed by the addition of HCl. This point can also

be seen in the effect of buffer capacity on the develop-
ment of the dark membrane potential due to the addition

of FeCl3 in symmetric acetate buffer solutions. As shown
in Table 2, a decrease in the buffer capacity of the
aqueous phases increased the magnitude of the dark
membrane potential, which could be developed by the
addition of a constant amount of FeCl3 in one of the
chambers. The adsorption of ferric ions onto the membrane
may also cause the development of the membrane potential.
This possibility, however, may be eliminated on the basis
of the observation that there was no extra effect following

the addition of FeCl into the aqueous phase other than

3
the previously mentioned pH effect observed due to the
addition of HCl. The measurements of the membrane
resistance also provided supporting evidence for the
above thought: there was no significant effect caused
by the presence of ferric ions. These observations now
clearly establish the fact that asymmetric addition of

FeCl3 induces the dark membrane potentials by generating

proton concentration gradient across the membrane.

53

TABLE 2.—-The develOpment of the dark membrane potential
as a function of the buffer concentration in the aqueous

 

 

phases.
Buffer Concentration Dark Membrane H Gradient
(M) Potential (mV) 9
0.1 - 1.8 i 0.1 <0.1
0.01 - 31 i 3 0.6
0.001 -110 110 1.9

 

Acetate buffer (pH 5.4)

FeCl (0.1 ml of 0.1 M) in the inside chamber

3

The active electrode is in the inside chamber.

54

The General Description of the
Photoresponses

 

 

The sign of the photoresponses depended upon the
location of FeCl3 as shown in Figure 5a. The direction
of the incident light did not influence the sign of the
responses. This can be seen in Figure 5b, which shows
the photoresponse recorded with the same electrode
arrangement as in Figure 5a, but with the position of
FeCl3 shifted to the outer chamber in order to simulate
the shift of the direction of the incident light with
respect to the preceding time course curve. It is worthy
to note that the shift of the direction of the incident
light did not change the sign, magnitude, or wave form
of the photoresponse.

Unlike the cases of FeCl3 and HCl, the concentration
gradient of KCl and NaCl did not show any effect on the
mode of the photoresponses. As shown in Figure 6, the
ten—fold concentration gradient of KCl had no effect on
the photoresponses. This fact is not unexpected since
the all-trans-retinal BLM is not sensitive to K+, Na+,
and Cl- ions. The agreement of the mode of the effect
of the salts on the photopotentials and on the dark
potentials strongly suggests that the main charge carriers,
or protons, in the dark state can interact with the
photoexcitation of the membrane selectively, or that

protons can function as charge carriers in observed

photopotentials.

55

Figure 5.--The effect of the position of FeCl with

3
respect to the incident light on the mode of
photoresponse. With FeCl3 (10“3 M) in the
inner chamber the photoresponse (for 1 second
illumination) shown in a was observed while

with FeCl (10.3 M) in the outer chamber the

3
photoresponse shown in b was recorded. Since
the outer electrode was grounded in both of
the cases, the data indicate that the photo-
responses were independent of the direction

of the illuminations and solely dependent on

the position of FeCl3.

 

 

 

 

lmv

 

60 Sec

57

Figure 6.--The effect of KCl concentration gradient
across bimolecular lipid membrane on the
photoresponses. The responses were recorded
with a single membrane in 10-3 M KCl (a) with
0.1 ml of 0.1 M FeCl3 in the inner chamber,
in 10.2 M KCl in the inner chamber (b), and
in 10—2 M KCl in both of the chambers (c).
The concentration of KCl in b and c were made
by injecting 0.1 ml of l M KCl, first to the
inner chamber, and then to the outer chamber.

The illumination period was one second for

each recording.

 

 

“ c-
7 5? i
m<
(D
n

59

Due to the close similarity between the biphasic
mode of the photoresponses of all-Epapg-retinal BLM
and that of the early receptor potentials of photoreceptors,
the convention of assigning R1 and R2 for the fast component
and the slow component respectively was adapted in the
present work.

The effect of the duration of illumination was
investigated with 0.1 M KCl solution with 10.3 M FeCl3 in
the inner chamber. Shown in Figure 7, a short time
illumination had only the R1 component and prolonging the
illumination generated the R2 phase while R1 reached the
apparent maximum value. The duration of the illumination
for the appearance and disappearance was dependent on the

temperature as shown in Table 3.

The illumination time of one second for the present

 

experiments was arbitrarily selected: at one second
illumination the saturation of the R1 component did occur
along with the reasonably high R2 component. Indeed, the
prolonged illumination caused irreversible changes on the
membrane potential to such an extent that duration of
illumination longer than one second was not desirable:
the recovery of the initial membrane potentials was not
observed as shown in Figures 7 and 8.

The selection of all-trans-retinal as the visual
pigment of the present membrane model was made partly
because rhodOpsin had been proved to have photoregeneration,

which would complicate the mode of the photoresponses. In

60

Figure 7.--The effect of the duration of illumination
on the mode of photoresponses. The
illumination times were indicated in
seconds. The photoresponses were recorded
from a single membrane formed in 9.1 M KCl

3

solution with 10- M FeCl3 in the inner

chamber at 25°C.

lInV

605ec

 

 

 

 

 

 

62

TABLE 3.--The effect of temperature on the duration of
illumination for the appearance of R2 and the disappearance
of R1.

 

Duration of Illumination for
Temperature (°C)

 

 

R2 Appearance Rl Disappearance
(sec) (sec)
9.5 >1.0 -
16.0 0.5 >l.0
25.0 0.2 >1.0
36.0 0.04 >1.0
44.0 - <0.01

 

The membrane was formed in 0.1 M KCl solution and
FeCl was added to the inner chamber to make up 10’3 M
solu ion.

63

Figure 8.--The effect of repetitive illuminations on the

photoresponses. The letters shown on the

curves indicate times after the first illumina-

tion in minutes, as shown below. The duration

of each

a-

b—

illumination was 30 seconds.
0 minute

10

20

30

40

50

60

70

80

155

 

65

order to examine the regeneration problem a membrane was
repeatedly exposed to a long-time (30 second) illumination
and the dark membrane potential as well as the photo-
response was investigated. In Figure 8 the effect of
repetitive illuminations on the photoresponses is shown.
The photoresponse decreased in its magnitude with an
increasing number of illuminations, without changing the
wave form appreciably. Shown in Figure 9 the dark
potential also decreased with an increasing number of
illuminations. The decrease in the dark membrane potential
may indicate the generation of the photoproducts which
modify the microenvironment of the membrane. If so, the
decrease in the photoresponse should have a strong rela—
tionship with a decrease in the dark membrane potential,
since the photoproducts cannot be formed without the con-
sumption of the photosensitive pigment molecules, and the
presence of the products may decrease the rate of photo-
reaction via porduct inhibition. A close relationship
between the dark membrane potential and the photoresponse
is shown to exist in Figure 10. It is interesting that
the ratio of the photopotential to the dark potential
decreases as the number of illuminations increases. This
decrease in the efficiency of the dark potential for
generation of the photOpotential in an increasing number
of illuminations, and the apparent lack of changes in the

mode of the time course of the photoresponse with an

66

Figure 9.--Photopotential (R2) and dark membrane
potential as a function of time from

first illumination.

the

:5.

32:30; 3.95562 foo

0

~80

 

 

o

2
.
a.

4
.
—

d-OO

 

+40,

+30 ..

+ 20?

.25

+10h

.2230; 29.:

 

9O

60

30

(min)

Time

68

Figurs 10.—-The ratio of photOpotential (R2) to the
dark membrane potential as a function of

the time from the first illumination.

 

 

00

120

 

6.0 —

Brenton foo

 

_otcm.oao.o.._m

2.0 “

 

Time (min.)

70

increasing number Of illuminations, and the apparent lack
of changes in the mode of the time course of the photo-
response with an increasing number of illuminations
clearly demonstrate that the regeneration problem does
not exist in the present model system. The irreversible
changes in the membrane potential and photoresponse due
to the repetition of long-term illuminations may suggest
the possible existence of the photoreaction which leads
to the production of membrane active chemical species,
possibly protons.

Light Intensity Dependence of
the Photoresponses

 

 

Due to the fact that the photoresponses could be
influenced by temperature and the dark membrane potential
could be controlled by pH gradient across the membrane,
the mode of the intensity dependence of the photo-
response could also be dependent on the temperature and
the pH gradient. In order to examine this point, the
intensity dependence was studied at various pH gradients
as a function of temperature.

The data shown in Figures 11-18 were taken by making
a membrane at a given temperature, varying the pH gradient,
and measuring the intensity dependence of the photo-
responses at each pH gradient. The inner chamber contained

-3 3

10 M FeCl3 in 10- M KCl solution and the pH of the outer

chamber was varied by the addition of HCl to the original

71

Figure ll.--The effect of relative intensity of light
on R1 component at ApH 0.3 as a function of
temperature. The inner pH was 3.3 and the

outer pH was higher. The inner chamber

3 M FeCl in 10"3 M KCl

3

while the outer chanber contained HCl and

contained about 10-

the same concentration of KCl.

 

RI (mV)

 

 

 

 

 

 

Relative Light Intensity

1’:
o
C
C
. A
o 25°cfi‘
L.
o '*<
0 15C
.
10°C
I l l l
.2 .4 .6 .8 1.0

73

Figure 12.--The effect of relative intensity of light on

R1 component at ApH 0.6 as a function of
temperature. The inner pH was 3.3 and the

outer pH was higher. The inner chamber con-

tained about 10-3 M FeCl3 in 10’3 M KCl

while the outer chamber contained HCl and

the same concentration of KCl.

("'V)

R1

 

 

 

 

 

 

 

/
25°C.
0
J
15°C
0
10°C
1 l l
.4 6 - 8' 1.0

Relative Light Intensity

75

Figure l3.--The effect of relative intensity of light on

Rl component at ApH 1.0 as a function of
temperature. The inner pH was 3.3 and the

outer pH was higher. The inner chamber

contained about 10'.3 M FeCl3 in 10-3 M KCl

while the outer chamber contained HCl and

the same concentration of KCl.

R. (mv)

 

 

 

 

 

 

 

 

251: T'
15°C
#4
10°C
l.-
l l l l
.2 -4 .6 .3 1.0

Relative Light Intensity

77

Figure l4.-—The effect of relative intensity of light
on Rl component at ApH 2.0 as a function
of temperature. The inner pH was 3.3 and
the outer pH was higher. The inner chamber
contained about 10.3 M FeCl3 in 10'.3 M KCl
while the outer chamber contained HCl and the

same concentration of KCl.

 

 

 

 

 

-1 l—
0
15C
-2
3‘
E
: O
a:
—3
10°C
-4
*5 l 1 l
.4 6 .8

Relative Light Intensity

LO

79

Figure 15.--The effect of relative intensity of light on

R2/Rl at ApH 0.3 as a function of temperature.
The inner pH was 3.3 and the outer pH was

higher. The inner chamber contained about

10—3 M FeCl3 in 10-3 M KCl while the outer

chamber contained HCl and the same concen-

tration of KCl.

 

R. .
[/R‘

 

 

 

 

 

 

 

2-0
1.5 _
,_l.o b.
M
N\
M
0
25C
O-Sr
15°C
0
10°C
0 l ”:31 ‘70 1 J
O .2 4 O .8

Relative Light Intensity

 

81

Figure l6.--The effect of relative intensity of light
on R2/Rl at ApH 0.6 as a function of
temperature. The inner pH was 3.3 and the
outer pH was higher. The inner chamber
contained about 10_3 M FeCl3 in 10-3 M KCl
while the outer chamber contained HCl and the

same concentration of KCl.

2.0

1.5

05*-

 

 

 

Relative Light Intensity

25C

 

l-O

83

Figure 17.--The effect of relative intensity of light
on R2/Rl at ApH 1.0 as a function of
temperature. The inner pH was 3.3 and the
outer pH was higher. The inner chamber
contained about 10.3 M FeCl3 in 10..3 M KCl

while the outer chamber contained HCl and

the same concentration of KCl.

2-0

 

1.5-

 

 

 

 

 

 

 

Relative Light Intensity

85

Figure 18.--The effect of relative intensity of light
on R2/Rl at ApH 2.0 as a function of
temperature. The inner pH was 3.3 and the

outer pH was higher. The inner chamber

3 M FeCl in 10"3 M KCl

3

while the outer chamber contained HCl and

contained about 10-

the same concentration of KCl.

R2
R1

 

2.0

l.5 #-

1.0 '5

 

 

 

25°C

15C

10C

 

 

 

Relative Light

Intensity

87

10-3 M KCl solution. R1 and the ratio of R2 to R1 were

plotted against the light intensity as a function of
temperature for a given pH gradient.

The intensity dependence of the R1 component can
be seen as a monotonously increasing curve in all cases.
At a proper pH gradient the ratio of R2 to R1 had a
constant value within a limited range Of the intensity
variation.

The effect of pH gradient across the membrane was
remarkable. The R1 component was almost insensitive to
a change in pH gradient while the R2 component increased
with an increase in pH gradient. This can be seen clearly
in the plot of R2/Rl against the light intensity where
the values increased generally with increasing pH
gradient (Figures 15-18).

Temperature dependence of R1 was evident, since at
lower temperatures higher photoresponses were observed.
The effect of temperature on the ratio of R2 to R1 could
be seen clearly: at higher temperatures the positive
peak increased with respect to the magnitude of the
negative peaks.

At lower pH gradient it can be seen that the genera-
tion of the R2 component requires higher intensity than
that of the R1 component. In order to observe the rela-
tionship more closely, the intensity dependence of the

photoresponses of the symmetric 0.1 M KCl system was

88

investigated. Shown in Figure 19, the magnitude of R1
and R2 could be fitted to a straight line in semilog plot
of the photopotential against the light intensity. The
threshold intensity was higher in the case of R2. On

the other hand, after its appearance the R2 component

had greater intensity dependence than the R1 component.

The Effect of Temperature
on the Photoresponses

 

 

All photoresponses were obtained at equilibrium
temperatures. The aqueous phases were symmetric with
respect to 0.1 M KCl and only the inner chamber contained
0.1 ml of 0.1 M FeCl3. The photoresponses for one
second illumination at various temperatures ranging from
9°C to 44°C are presented in Figure 20. At room tempera-
ture (25°C) the photoresponse showed a distinct biphasic
response. When the temperature was raised to 44°C
only R2 component could be seen. On the other hand the
lowering of temperatures down to 9°C abolished the R2
component completely and isolated the R1 component. It
is remarkable that this complete simulation of the
temperature dependence of the early receptor potential
could be achieved within this small range of temperature
variation.

The magnitude of R1 and R2 varied from membrane to
membrane, even at a given temperature, so much that a
direct comparison of the magnitudes of the photoresponses

at various temperatures could not be carried out

89

Figure l9.--The effect of light intensity on the
photopotential of all-Egaps-retinal
bimolecular lipid membrane. The membrane
was formed in 0.1 M KCl solution without
making any chemical assymetry. The
photopotentials were recorded from a

single membrane.

+.8

+36

."r
A

j-
to

PhotOpotenti al(mV)

.l
M

 

R2

 

O

 

 

 

 

Relative Light

Inten Sll‘y

91

Figure 20.--The time course of the photoresponse of
all-trans—retinal BLM. FeCl3 was placed
in the inner chamber. The duration of the
illumination was 1 second. The outside

pH was 5.4 and the inside pH was 3.3.

 

 

44°C ——

 

0
36C ‘1

25°C 1

15?:

 

 

9%

 

93

meaningfully. The ratio of the magnitude of R2 to that
of R1, however, was relatively free from membrane
variations and was the major concern. This ratio was
strongly dependent on temperature as shown in Figure 21.
It can be seen that the complete disappearance of R1
component occurred at 44°C and that of R2 at 9°C. R2
exhibited the same amplitude to R1 at a slightly lower
temperature than room temperature.

The general observations about the effect of
temperature on the mode of photoresponses were strengthened
by the studies carried out with single membranes subjected
for various temperatures. In this work it became possible
to relate the photopotential to the dark membrane
prOperties as a function of temperature.

The time courses of the photoresponses obtained with
the system which contained only HCl with pH gradient of
0.7 were shown in Figure 22. The same features observed
in Figure 20 can be seen but the temperatures for the
appearance of R2 and the disappearance of R1 were higher
than those of Figure 20. This difference may be due to
a smaller pH gradient across the membrane system shown
in Figure 22. Indeed as shown in the next section
increasing pH gradient could enhance the R2 component
significantly.

The data shown in Figure 23 were obtained in the

presence of ferric ions in the inner chamber while

94

Figure 21.--The ratio of R2 to R1 as a function of
temperature. The data shown in Figure 20

were used for the calculation of the

ratios.

 

:0
N

:0
—o

 

 

 

l5

25

T (°C)

35

 

96

Figure 22.--The time course of the photoresponses of
all-trans-retinal BLM as a function of
temperature: HCl system. The aqueous
phases contained only HCl with the inside
pH at 2.8 and outside pH at 3.5. The

duration of the illumination was 1 second.

 

_ _ ..____....._ -—

1 mV
10 Sec

 

 

 

 

 

 

 

 

 

 

26% j
15?: ‘1
ICC—1

46°C
40°C
3 2°C

98

Figure 23.--The time course of the photoresponses of
all-trans-retinal BLM as a function of
temperature: FeCl3 system. The aqueous
phases contained only HCl with the outside
pH at 3.5. The inside chamber contained,
in addition to HCl, FeCl3 to lower pH to

2.8. The duration of the illumination

was 1 second.

 

 

 

 

 

 

 

 

40C —_l *
25C -__l
e i
l0 Sec
15°C
'—1

 

 

100

maintaining the same pH gradient as the system used for
Obtaining the data presented in Figure 22. The magnitude
of the photoresponses was increased by approximately
four times. The temperature for the appearance of the
R2 component was elevated to about 40°C and the R1
component could be seen even at 47°C.

The dark membrane potentials of both HCl and FeCl3
systems increased with increasing temperature as shown
in Figure 24. The thermal coefficients of these two
systems were identical (Table 4). Due to the same
magnitude of the dark membrane potential development
and the identical thermal coefficient between the HCl
system and the FeCl3 system, it is very likely that the
differences of photoresponses between the two systems
may be attributed to the presence of ferric ions, which
are good electron acceptors.

The semilog plot of the R1 component and the dark
membrane resistance against the reciprocal temperature
(Figures 25 and 26) revealed the close relationship
between the two properties. Indeed, shown in Figure 27
the R1 component decreased with a decrease in the membrane
resistance while the R2 component increased. In this
particular experimental condition the thermal coefficients
of the dark membrane potential, dark membrane resistance,
and the R1 component were almost identical (Table 4). The

R2 components had relatively higher thermal coefficients.

101

Figure 24.--The temperature dependence of the dark
membrane potential of all-EEEEs—retinal
BLM in asymmetric HCl solution or
asymmetric FeCl3 solution.

a - The aqueous phase contained only HCl
with the inside pH at 2.8 and the
outside pH at 3.5. The thermal
coefficient obtained from the slope
of the straight line was 5.1 kcal/mole.

b - The aqueous phases contained only HCl
with the outside pH at 3.5 and the
inside chamber contained, in addition
to HCl, FeCl3 to lower pH to 2.8.

The thermal coefficient obtained from

the slope of the straight line was 4.9

kcal/mole.

 

 

 

‘O~~Q\s

 

l-l

_ _

36

34

32

 

 

100

80

60-

40—

r
0
8

100

35:02:30..

L
0
O

8500'.

_
0
4

20

1000

T<°C)

103

TABLE 4.--The thermal coefficients of hydrogen ion transfer
number, membrane resistance, and R1 and R2 components.

 

Thermal Coefficient

(kcal/mole) in

 

Parameter Observed

 

HCl FeCl3
Dark Membrane Potential 5.1 4.9
Membrane Resistance 6.9 6.9
R1 Component 9.1 6.9
R2 Component 18 15

 

104

Figure 25.--Dark membrane resistance and R1 photo—
potential as a function of temperature in
HCl system. See the legend of Figure 24

for the experimental details.

Rm - Membrane Resistance

Ri — Photopotential R1

 

 

 

 

A>EL .o.._.ca.oao.o:m E
2
8 I j
d 1 it: u _ 4 -

O

L
O

m
l
m l

R

.1
J
7ll—r bl 7—l-lb - - lP 6
O O O
X .NEu.E;o.0uco:_uo¢ mu
8 2

3.6

3.4

3.2

1000

T (°K)

106

Figure 26.--Dark membrane resistance and R1
photopotential as a function of
temperature in FeCl3 system. See the

legend of Figure 24 for the experimental

details.

Rm - Membrane Resistance

R1 — Photopotential R1

 

80

:51 3:58.32... z.

0.2
3.6

 

 

8X10

 

A Eu .E gov cue-.33”.

N

 

°l

2X10

3-5

3.3 3.4
1000

3-2

34

T(°K)

108

Figure 27.--The relationship between two components of
photopotentials and dark membrane resistance.
The experimental conditions were the same as

those shown in the legend of Figure 24a.

 

 

R2

 

Pho’Opotentu al (mV)

 

 

 

 

Resistance ( lO7Ohm .cm2 )

110

The Effect of pH on the Photoresponses

The fact that resting membrane potential was solely
dependent on the gradient of proton concentrations in the
bathing solutions separated by the membrane immediately
suggests that the mode of photoresponse should show sig-
nificant variations upon changes in pH gradient. Further-
more, the possible generation of protons as one of the
photoproducts may complicate the effect of pH on the
photoresponses.

When pH of the outer chamber was increased by the
addition of KOH, the magnitude of the slow component (R2)
increased, and masking of the negative fast component
(R1) could be seen unambiguously (Figure 28). The magni-
tude of R2 increased with an increase in pH in a linear
relationship to pH as shown in Figure 29.

The Effect of Buffer Capacity
on the Photoresponses

 

 

Due to the essential role of the hydrogen ion
concentration gradient in determining the mode of the
photoresponses, the effect of buffer capacity was also
investigated.

In order to study the effect Of buffer capacity
the proton concentration gradient has to be kept constant
for various buffer capacities. This was best achieved
with the use of symmetric KCl systems and symmetric

acetate buffer systems. As shown in Figure 30, changes

111

Figure 28.--The effect of pH on the photoresponses. The
values of pH were those of the outer chamber
adjusted with KOH and HCl in the 0.1 M KCl
solution which was also contained in the

inner chamber (pH 5.4).

 

5-4

.1 mV

l20$ec

 

6.6

 

7.4

 

8.0

 

8.4

113

Figure 29.-~Rl and R2 as a function of pH gradient
across the membrane. Without the presence
of a strong electron acceptor both R1 and R2
are linearly dependent on the pH gradient.

The data used were those of Figure 28.

 

 

 

 

 

 

125 J

0.2

3.0

2.0

1.0

APH

106

Figure 26.—-Dark membrane resistance and R1
photopotential as a function of
temperature in FeCl3 system. See the

legend of Figure 24 for the experimental

details.

Rm - Membrane Resistance

R1 — PhotOpotential R1

80

$5.. 3.22822... 2.

0.2
3 6

 

 

8X10

 

.NEu . E 3 av cue—wigs

1
3.3 3-5

1000

l

34

 

2x10I

T(°K)

108

Figure 27.--The relationship between two components Of
photOpotentials and dark membrane resistance.
The experimental conditions were the same as

those shown in the legend of Figure 24a.

 

 

PhotOpotentu al (mV)

 

+.4 ~

+
L:
1

R2

 

 

.. tel—l—H—|__.__._J__._J

'708 09‘

2

Resistance (10

3 4

7ohm.cm

 

 

2)

5

6

 

7

110

The Effect of pH on the Photoresponses

 

The fact that resting membrane potential was solely
dependent on the gradient of proton concentrations in the
bathing solutions separated by the membrane immediately
suggests that the mode of photoresponse should show sig-
nificant variations upon changes in pH gradient. Further-
more, the possible generation of protons as one of the
photoproducts may complicate the effect of pH on the
photoresponses.

When pH of the outer chamber was increased by the
addition of KOH, the magnitude of the slow component (R2)
increased, and masking of the negative fast component
(R1) could be seen unambiguously (Figure 28). The magni-
tude of R2 increased with an increase in pH in a linear
relationship to pH as shown in Figure 29.

The Effect of Buffer Capacity
on the Photoresponses

 

 

Due to the essential role of the hydrogen ion
concentration gradient in determining the mode of the
photoresponses, the effect Of buffer capacity was also
investigated.

In order to study the effect of buffer capacity
the proton concentration gradient has to be kept constant
for various buffer capacities. This was best achieved
with the use of symmetric KCl systems and symmetric

acetate buffer systems. As shown in Figure 30, changes

111

Figure 28.-—The effect of pH on the photoresponses. The
values of pH were those of the outer chamber
adjusted with KOH and HCl in the 0.1 M KCl
solution which was also contained in the

inner chamber (pH 5.4).

 

 

8.4

8.0

7.4

6~6

5-4

113

Figure 29.——Rl and R2 as a function of pH gradient
across the membrane. Without the presence
of a strong electron acceptor both R1 and R2
are linearly dependent on the pH gradient.

The data used were those of Figure 28.

 

 

 

 

 

 

“>5. mm

0.1-

:5 c.

0.2

3.0

2.0

1.0

ApH

115

Figure 30.--The time course of the photoresponses of

all-trans—retinal BLM in symmetric KCl

 

solutions as a function of salt concentration.
The upward arrow indicates the light-on and
the downward arrow the light-off for 1 second

illumination.

 

10" M

I 0"‘M

ll

117

in the concentration of KCl had no essential effect on

the mode of photoresponses. In acetate buffer systems,
however, the concentration effect was remarkable (Figure
31). In a 10.3 M system R1 and R2 could be seen in
exactly similar mode to those Of KCl systems. Indeed,

at this buffer concentration there was almost no buffering
action as demonstrated in Table 2. In a 10.2 M acetate
buffer system the R2 component was completely eliminated
and the R1 component was slightly enhanced. As the con-
centration was increased to 10.1 M the R1 component was
essentially unaffected. These features are shown in
Figure 32. In order to separate the effect of buffering
capacity from that of the ionic strength the concentra-
tion dependence of the dark membrane resistance was comm
pared with the concentration dependence of the photo-
response in KCl systems and acetate buffer systems as
shown in Figure 32. In this comparison it can be clearly
seen that the buffer capacity influenced mainly the R2
component so that in the presence of 10-2 M acetate buffer
the complete elimination of the component was Observed.
The fact that this is the effect of buffer capacity can

be recognized in the observation that the concentration
dependence Of the membrane resistance in acetate buffer
systems was identical to the KCl systems, in which R2

was essentially unaffected. The effect of buffer capacity

on the R2 component is strong supporting evidence for the

118

Figure 31.——The time course of the photoresponses of
all-trans-retinal BLM in symmetric acetate
buffer solutions as a function of buffer
concentration. The upward arrow indicates
the light-on and the downward arrow the

light-off for 1 second illumination.

 

 

120

Figure 32.--The magnitude of photoresponses and dark
membrane resistance as a function of the
concentration of KCl or sodium acetate

buffer systems.

R1 - R1 Photoresponse
R2 - R2 Photoresponse

Rm - Dark membrane resistance

$2.32 $2.: 2.

 

 

 

 

 

Log]0 (Concentration)

 

 

 

 

 

 

 

q a a _ _ 1— d
.L
r
l e '
l H
C u
K B
I m
C C.
2 .l 2 K to
R m R R B
. h _ _ _ p _
2 .l 0 3 2 .l. 0 6 4

.>EL Z. A>Eu:~_ ANEu.E;Mo: Ea

122

involvement of the hydrogen ion in generation of the R2
component.

An extremely interesting observation was made on a
long-time illumination. In 0.1 M KCl solution with 10—3 M
FeCl3 in the inner chamber and at zero pH gradient, a 60
second illumination gave triphasic responses which were
apparently the modified wave forms of the biphasic
responses as shown in Figure 33a. The R1 response was
taken over by the R2 component which did not pass over
the resting membrane potential developed before the
illumination. Instead the R2 component bounced back and
develOped in the same direction as the R1 component.

With a slower rate than R1, this component develOped con-
tinuously up to the light-off, after which the potential
decreased instantaneously to the base line, and then
slowly developed the usual R2 component passing over the
base line. The final resting potential was approximately
the same magnitude to the inverse R2 component but in the
Opposite direction. This base line change was not
reversible.

In the case of high buffer capacity, shown in
Figure 33b, the photoresponse was dramatically different
from that of Figure 33a. The fast developing R1 was
reversed to the base line like the response shown in
Figure 33a. After this recovery of the original potential,
the reversed R2 component developed very slowly and in a

more suppressed manner than that of Figure 33a. Upon the

123

Figure 33.--The time course of photoresponses with 60
second illuminations at pH 3.3 and ApH 0.
The downward arrow indicates the light-on

while the upward arrow indicates the light-off.

a - 0.1 M KCl solution: the inner chamber

3

contained 10- M FeCl and pH of the

3

outer chamber was lowered with HCl.

b. 0.1 M sodium acetate buffer solution:

the inner chamber contained 10-3 M FeCl3.

125

light-off the response slowly recovered the original
base line and never passed over it, unlike the response
shown in Figure 33a. These two types of photoresponses
following a long time illumination are very important in
elucidating the mechanism of generation of the photo-

potentials.

CHAPTER VI

DISCUSSION

General Discussion

 

In this section pertinent observations will be
discussed to establish an overall view of the physical
processes leading to the generation of the photoresponses.
Especially, the reasoning for the assignment of hole
diffusion potential for the R1 component and proton dif-
fusion potential for the R2 component will be presented
after establishing the mechanism of the generation of the
dark membrane potentials. A detailed quantitative theory
will be given in the next section, followed by the
experimental verification of the theory by comparing the
theoretical predictions on the mode of the photoresponses
to the experimental observations carried out in the present
work.

The ionic transference numbers of KCl, NaCl, and HCl
show that all-Epapg-retinal BLM is cation selective and
the development of the dark membrane potentials can be
explained with proton transport alone. When HCl was added
in the inner chamber, the negative potential at the inner
electrode developed. This may be interpreted as the proton

diffusion toward the outer surface leaving chloride ions

126

127

in the inner side. As to the question of whether the
observed protonic potential is actually due to diffusion
or adsorption the answer can be given safely as diffusion,
since the Nernst equation can hold. If the potential
were due to adsorption, the Observed negative potential
should be due to chloride ions but in the presence of
the concentration gradient of KCl or NaCl in two aqueous
phases such a large negative potential was not observed.
The elucidation of the mechanism by which this
protonic potential develops is certainly interesting.
In this work, however, the question was not pursued
further since (1) other evidence as presented below
indicated that protonic conduction increased with water
content in the hydrocarbon layer in BLM, suggesting the
possibility of the contribution of water pores to the
transport of protons, and (2) the temperature dependence
of the membrane potential and resistance was not influenced
by the presence of electron acceptors such as ferric ions,
suggesting that the potential development observed by
creating pH gradient across BLM may be due to the simple
diffusion of protons without performing other functions
such as catalyzing the electronic charge injection, as
reported in the case of anthracene/water interfaces
(Beyerle and Gerischer, 1967). The presence of the proton
transport channel separated from the electronic conduction

channel which is most likely to be a conjugated double

128

bond of all—Egaps-retinal is assumed to exist in the
model membranes used in the present work.

From the sign of the R1 component it is reasonable
to consider that this component is likely to be either the
diffusion of positive charge carriers (protons and/or
holes) to the outer surface and/or the diffusion of
electrons to the inner surface. The possibility that R1
may be mediated by protons can be eliminated by the
following Observations:

1. The sign of R1 was totally dependent on the

position of the electron acceptors.

2. The magnitude could be greatly enhanced, as
much as 103 times, by the addition of electron
acceptors.

3. In the presence of the high concentration of
electron acceptors the magnitude of R1 was
independent of the proton concentration
gradient across the membrane.

4. At higher buffer capacity, where the R2
component was completely abolished, the R1
component could still be observed.

Thus, our attention can be focused to electrons and holes
as the charge carriers of the R1 component.

As prOposed by Tien (1968a, b) in explaining the
photopotential of chlorophyll BLM, generation of R1 may
.be explained as follows: due to the extremely thin

structure generation of electrons and holes is symmetric

129

if there is no chemical asymmetry. On the other hand,

in the presence of electron acceptors at the inner surface
the excited electrons may tend to be withdrawn toward the

inner surface, generating a higher concentration of holes

in the bulk of hydrocarbon or at the outer surface.

An immediate establishment of the strong electric
field across the membrane may have an effect on the space
charge distribution across the membrane. The charge
carriers which can be influenced by this photoinduced
field are protons, since electrons are trapped by ferric
ions, the membrane is impermeable to chloride ions, and
holes are the sources of the induced field. Thus, the R2
component may be considered as the potential developed
by the diffusion of protons toward the inner surface, or
anode.

In accord to the above expectation, the R2 component
increases with increasing temperature and pH gradient. An
increase in temperature influences the R2 component by
increasing protonic conductance or increasing protonic
mobility whereas the pH gradient increases the magnitude
of R2 by increasing the protonic concentration at the
outer surface.

In addition to the possible interaction of protonic
space charges in the generation of the R2 component, there
seems to be another possibility. The accumulation of holes
at the outer surface may generate protons via the oxida-

tion of water molecules as proposed by Tien (1968a, b) in

130

the case of chlorophyll BLM. The newly generated protons
contributing to pH gradient will cause a decrease in the
protonic membrane potentials established at the dark
conditions. Indeed, the effect of a strong buffer
capacity in abolishing the R2 component cannot be
explained without taking into account this photogeneration
of protons. The mechanism of the photogeneration of
protons is still unknown not only in the present model
system but also in well studied chloroplast systems.
Nevertheless, since the peak wavelength Of the absorption
spectrum of all—ppapg-retinal has 80 kcal/einstein and the
bond energy of aldehyde-water hydrogen bonds is 2 - 3
kcal/mole (Karyakin, Muradova, Sayenko, 1969), the photo-
generation of protons is indeed possible energetically.
Since the membrane is impermeable to chloride and
ferric ions, the recovery of the original dark potentials
can be achieved by restoration of the chemical equilibrium.
Specifically, the dissociation of HCl salts to protons and
chloride ions and the reestablishment of equilibrium
between ferric ions and ferrous ions may generate protons
in the inner chamber and, later, the diffusion of protons
will reestablish the original membrane potentials. For a
long-time illumination the recovery of the membrane
potentials did not take place. This may be due to the
disturbance of the electric double layers to such an

extent that the distribution of chloride and ferric ions

131

is partially modified from that of the dark state due

to the prolonged illuminations.

Theory

In order to set up kinetic equations describing the
development of the biphasic photoresponses, it is
necessary to establish the physical model Of the seem-
ingly complex phenomena. The present model is based on
the two basic principles which were confirmed in this
work. The first principle is that the membrane is
permeable to protons, but not to the other ions studied
(e.g., K+ and Na+). The second principle is that any
external force which disturbs the existing electro-
chemical equilibrium has to be acted upon by a counter
force to maintain microsc0pic electrochemical neutrality.
Based on these principles, it is proposed that the
initial electronic phenomenon following light absorption
can generate the ionic charge transport, whose major charge
carriers are protons.

In this work the resting membrane potential was
expressed as the potential of the active electrode which
was placed in the inner chamber. The photopotential was

expressed as R1 and R2, which were defined as:

R1=V. -v (3)

R2=V -v (4)

132

where V .
m

1n' V , and V0 were the value of the membrane

max
potentials at the first peak, the second peak and the
resting state, with signs taken in the sense of the
inside potential minus the outside potential.

All time courses of the photoresponses could be

expressed with the following two equations:

v = - A(1 - e- at) + B(l - e- bt) (5)
for 0 i t 3 t0 (6)
and
V = — A(1 - e- ato)e_ a(t - to)
+ B(l _ e- bto)e- b(t - to) (7)
for t0 1 t (8)

where A, B, a, and b are constants and to is the time (t)
for the light—off taking t = 0 for the light-on. All
constants are functions of various experimental parameters
and the assignment of numerical values to them requires
to specify the exact conditions under which the constants
have been evaluated. Because of this it is desirable to
evaluate the constants in terms of experimentally measurable
parameters.

The sign of A is negative to show that the first term
expresses the negative or the major component of R1 while

the second term expresses the positive or the major

133

component of R2. In spite of the fact that the Observed
photoresponses are the results of redox reactions and

the diffusion of the products of the redox reactions or
ionic charge carriers, the time courses of the photo-
responses are determined with two terms of exponential
functions. This implies either that the time constant of
each component is the sum of the redox reactions and
diffusion terms or that it is a pure time constant of

either the redox reactions or diffusion processes:

(9)

a=a +8.... .
redox diffu51on

b = bredox + bdiffusion (10)
or

a = aredox (11)

b = bredox (12)
or

a = adiffusion (13)

b = bdiffusion (14)

Due to the fact that these time constants (a and b)
are far greater than any of the anticipated rate con-
stants of redox reactions and diffusion processes of holes
and protons, it is more fruitful to analyze the kinetic

equation in terms of electric circuit theory.

134

The forementioned two terms of exponential functions
are best expressed as the representations of the time
courses of two RC circuits. Shown in Figure 34, the hole
component may be represented with circuit 1 and the proton
component with circuit 2. Then, the following equations

can be written down:

1 _ - G t/C
VC VO — (Vl VO)(1 e 1 m) (15)
2 _ _ _ _ - G t/C
VC VO — (V2 VO)(1 e 2 m) (16)
where
Vi = capacitative potential of hole component
v: = capacitative potential of proton component
VO = dark potential
Vl = saturation potential of hole component
V2 = saturation potential of proton component
G1 = conductance of hole component
G2 = conductance of proton component
Cm = membrane capacitance

The combined system shown in Figure 34c can be

expressed as:

135

Figure 34.--The equivalent circuits for the generation

of photoresponses of all-trans-retinal BLM.

a - hole system designated with subscript 1

b - proton system designated with subscript 2

c — combined system
G - membrane conductance
V - membrane potential

C - membrane capacitance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

137

total _ l _ _ -G t/C
Vc VO — G + G (Vl VO)(1 e 1 m)
1 2
1 ____.__°2 (v — v )(1 — e’Gzt/Cm) (17)
G + G 2 o

l 2

total . . . .

where VC 15 the total capac1tat1ve potential. The

Equation (17) expresses the time course of the membrane
potential for 0 i t i to' After the light-off, t0 3 t,

the time course of the photopotential can be expressed by:

G

total _ _ 1 _ _ -G t /C -G (t-t )/C
VC VO — 5——:—E—(Vl VO)(1 e l o m)e l o m
l 2
G -Gt/C _ _
+ ————g——(V -V )(1-e 2 O m)e G2(t to)/Cm (18)
G1 + G2 2 0

Equations (17) and (18) have the same properties of

Equations (5) and (7), provided:

Gl
A = ——————— (V - V ) (19)
G1 + G2 1 0
G2
B = (V - V ) (20)
G1 + G2 2 O
a = Gl/Cm (21)

b = Gz/C (22)

138

Our task is now to express V1 and V2 in terms of
other physical parameters. Since these potentials are
saturation potentials, by putting t = w in Equations (15)

and (16) we get:

1 _
VC — V1 (23)
and
2
Vc — V2 (24)

Referring to Figure 35 it is possible to evaluate
saturation membrane potentials in terms of the saturation

concentrations of charge carriers:

Vphoto vphoto Vdark

A 1 1 ‘ 1

 

= Vl - Vo
Ci,h
_ RT i,h i,h _ o,h _ t
_ F {[ln(Cp + Ct ) lnCt ] 1n Co'h}
t
Ci,h
zfllnL—
F cl'h (25)
t
h h

where c;' and C3’ are the saturation concentrations of

photogenerated holes in the inside surface and outside

:’h and Ci’h are the satura-

tion concentrations of thermally generated holes in the

surface, respectively, and C

inner surface and outer surface, respectively.

139

Figure 35.-~The analysis of two components of

photopotentials.
VO - resting membrane potential

V - saturation hole potential

1

V2 - saturation proton potential

141

Similarly we obtain:

photo _ photo _ dark
AV2 — V2 V2
= V2 - Vo

 

Co,H
_ RT O,H O,H _ i,H d
— F {[ln(C + Cd ) lnCd ] lnCi,H}
d
CO,H
= KEML (26)
F Co,H
d

where C's now stand for proton concentrations at the
saturation state.

The terms expressing the concentrations must be
represented by more basic parameters. In order to do this
the establishment of the whole photochemical reaction
starting from the absorption of light to the restoration
of the original resting state of electrochemical
equilibrium must be carried out. Such an attempt must
introduce a great degree of simplification due to the
many parameters involved in the reaction sequence, whose
mathematical treatment could be unduly complicated for the
prOper evaluation of the experimental observations.
However, there is a risk in attempting such a simplifica-
tion as it may result in an unrealistic description.

The type of primary excitation species was not

determined in the present work; thus, either excitons or

142

excited electrons can serve well for the present
discussion as stated below. Whether the excited species
is singlet or triplet could not be decided experimentally,
but from the time range observed in the present work the
species with a longer lifetime is more likely to be
involved. The triplet state has a lifetime on the order
of one second in the case of carotenoid glasses (Rosenberg,
1958) and less than 10"5 second in the case of various
organic and aqueous solutions of retinals (Abrahamson,
Adams, Wulff, 1959). The longer lifetime of the triplet
state in solid is likely to be due to the presence of
traps, and the similar possibility may exist in the case
of BLM since the pigment molecules are arranged in a
regular structure (Tien, 1967). Thus, it is assumed that
the primary excitation step is the generation of triplet
excitons.

The rate of exciton generation can be expressed by:
e

C

r
___ = Cr€I e-eC l
o

Q:

e l
- C (F; + k1) (27)

Q:

where Ce is exciton concentration, Cr is retinal con-
centration, s is molar extinction coefficient, I0 is the
intensity of the incident light, 1 is the thickness of the
membrane, T is the lifetime of excitons, and k1 is the

e

rate constant of hole generation.

143
At equilibrium:
——— = 0 (28)

Then, from Equation (27) we get:

r
e Cr'reefloe-EC l
C = 1 + k T (29)
1 e

 

In deriving Equations (27) and (29) it was assumed
that the reaction kinetics of the generation of charge
carriers from excitons as the first order. This simpli-
fication may be justified as shown in the development of
the possible mechanism of charge carrier generation as
described in the following paragraphs.

The processes leading to the generation of electronic
charge carriers can be considered in the following two
ways. The first possibility is the charge separation via
the photogenerated excitons and second possibility is
the charge transfer complex formation between excited
retinal and electron acceptors or electron donors. The
first pathway may be written as:

k k
e+A-—-——9——->Ae——C—-—)A-+(B (30)

b

144

k' k'
e+D——°‘———>De———g——)D++e (31)
W—
kb

where e is exciton, A is electron acceptor, D is electron
donor, Ae and De are complexes for acceptor—exciton and
donor-exciton, respectively (these complexes may be
considered as surface traps), O is hole, e- is electron,
and k's are rate constants for respective processes. For

the charge transfer complex pathway reactions may be

written as:

R +A—————-——-=-PAR c A +R (32)

R+D——————-¥DR c D+R- (33)

where R* is the excited retinal molecule.

Between these two types of pathways there are
similarities. Excited species form a complex with
acceptor or donor reversibly and the product formation is
irreversible. Donors and acceptors are fixed at the
interfaces and can be considered as fixed sites for
adsorption. Excitons and excited retinals can be con-
sidered as the species colliding onto the fixed sites.

It is assumed that excitons and excited retinals can

145

interact with the adsorption sites in the fashion of a
random walk. If the above assumptions are correct, then,
the kinetics can be studied with the Langmuir adsorption
isotherm. Before the development of the kinetic equa-
tions using the Langmuir adsorption isotherm, it is
necessary to clarify the details of the present model.

Because of the uncertainty on the possible pathway
between the proposed two pathways, the rest of the dis-
cussion will be focused on the exciton pathway. This
selection does not create any inconvenience since the
other pathway is quite similar to the former one.

It is expected that water itself can perform the
role of electron acceptor as well as electron donor in
the absence of better acceptors or donors. The reactions
taking place at the inside surface are the summation of
the reactions expressed by Equations (30) and (31). The
observed net reaction is, then, the difference between
the sum of the similar reactions occurring at the outside
surface and the sum of the similar reactions at the inside
surface. In the case of a symmetric system the rate of
the generation of charge carriers is the same between two
surfaces, thus, we should expect that potential develop-
ment is negligible. Indeed, this is the case as commonly
observed in the present work. Because of the complexity
in treating the total reactions as a general case, and

yet the considerable generality of the single reaction,

146

treated in the following paragraphs will be an asymmetric
case with ferric ions in the inside chamber. This
selection could practically fix the sign of photoresponses
and enhance the magnitude, indicating such a simplified
treatment is not only desirable but also required. Due
to this factor of asymmetry a great simplification can
be made in discussing the kinetics of the development of
the electronic component.

In the simplest situation ferric ions function as
the sites and excitons adsorb onto the surface sites.

The reversible adsorption can be expressed by Langmuir's

isotherm (1918). Let s be the fraction Of surface that
is covered and 1 - s the fraction that is bare. The rate
of adsorption is then kaC(l - s), where C is the concen-

tration of the adsorbate (exciton) and ka a constant
shown in Equation (30) and the rate of desorption is

kbs. At equilibrium the rates are equal, or:

 

l_S=-];-°-C=Kc (34)
b

where K, equal to ka/kb, is a constant. This equation

can be written as:

_ KC
S — 177::- ‘35)

The process of charge separation may be regarded
as the process involving the following four consecutive

steps. These are:

147

l. diffusion of the exciton to the surface

2. adsorption of the exciton onto the surface

3. reaction on the surface

4. desorption of the product (hole).

It is assumed that processes 2 through 4 are con-
sidered as the rate determining processes as a unit, as
treated by Langmuir (1921) in deriving his kinetic
equation. This assumption was made since the activation
energy may be required for adsorption and desorption.
These two processes cannot be separated experimentally,
so it is convenient to consider that these steps function
as a unit. The rate of reaction (v) is, then, prOportional

to s and may therefore be written as:

<1
ll

7:

m

(36)

where K. is the prOportionality constant. This deriva-
tion was based on the assumption that the adsorption
equilibrium was not disturbed by the occurrence of the
reaction.

At sufficiently high reactant concentration the
rate is independent of the concentration, which means
that the kinetics are zero order. Under these conditions
KC >>l, so that Equation (36) reduces to v = K'. At low

reactant concentrations, which are the range of concern in

the present work, when KC <<1, Equation (36) reduces to:

148

v = K KC (37)

and the kinetics are first order.

Similarly a first order rate equation can be derived
for proton generation at the opposite surface of the BLM
by assuming the Langmuir type adsorption of holes at the
surface sites as an intermediate step. The diffusion
process of holes and protons can be treated as the first
order rate reactions. It is also possible to specify
that all reactions are biased forwardly and backward rate
constants are negligible. From the above considerations
it becomes possible to construct the reaction flow
diagram composed with first order irreversible reactions
as shown in Figure 36. The rate of the reactions of this

system shown in Figure 36 can be written as follows:

 

dC _ _ r
dt I koC (38)
dCe r e
1 h
dC ’ .
p _ e _ 1,h o,h
dt _ klc (k4+ k5 + D1)Cp + 03cp (40)
o,h
dc = k Ce - (k + k + D )co'h + D ci'h (41)
dt 6 2 7 3 p l p

149

Figure 36.--The photochemical reactions and diffusion
processes involving excitons (e), holes (h),
and hydrogen ions (H). k's are reaction
rate constants and D's are diffusion
coefficients. r is all—trans-retinal.

L is hydrocarbon layer, 0 is outside
aqueous phase, and i is inside aqueous

phase.

dC

dt

dCi,H
dt

dcHCl

dt

151

- (k8 + 02)c;'H + D4C;'H (42)
- (k3 + k9)Cl’H + chg'H (43)
(44)

For the sake of simplicity without losing the validity

of the discussion,

as shown in

 

 

Figure

let k

37.

4 = k

Then,

5

=k6=k7=k8:k9

the above equations become:

= O

(45)

(46)

(47)

(48)

(49)

(50)

152

Figure 37.-~Simplified photochemical reactions and

diffusion processes.

hu

 

 

 

 

 

\/

 

154

dcHCl

1
dt 3 p (51)

The reaction scheme shown above may be expressed as
the reactions of most possible reactants (ferric ions as
acceptors and water molecules as donors) as shown in
Figure 38. Shown in the diagram are four major processes:
(1) photon absorption leading to the generation of
excitons, then holes, (2) the diffusion of holes across
the hydrocarbon layer of BLM in the same sense of the
dark current, (3) interaction of holes with water
molecules leading to the proton generation, and (4) the
back flow of protons due to the newly formed proton or
disturbance of the protonic concentration gradient across
BLM.

Upon establishing the simplified reaction scheme and
the rate equations the equilibrium concentrations of the
terms shown in Equations (25) and (26) can now be

evaluated:

dcl'h e Epjh i h
_ _ __ _ I
dt — le Th Dle (52)

where Th is the hole lifetime.

At equilibrium:

= O (53)

dC
dt

155

Figure 38.--Possible redox reactions at the membrane

surfaces.

 

 

 

 

 

o L
hu + r—e r"
r’—> F +6
5 e+A:=Ae
Ae—>A’+e
l
XMDZDO +
DO—>D+ > D +A->DA
.—> e+F§i3CIEz
Fe3i3CI-ie --9
Fé‘éo’ +01?
\V e
$+H20—
l + + ++ -
402 H e H Cl->HC|

 

 

157

 

 

 

and
r
. Crk T T €I e-EC l
C1,h = f l e h (54)
p (l + lee)(1 + tthY
Similarly the rate of photogeneration of protons can
be expressed as (Equation 49):
O H
dC ’
p = o,h _ o,H
at kZCp DZCP (55)
where
r
r -€C l
Co'h = C klkzrerhsIoe (56)
p (I+_klte)(l + thl)(1 + kzth)
At equilibrium dcg’H/dt = 0 and the substitution of
Equation (56) into Equation (55) gives:
r
r -€C 1
CO'H = C k1k2TeTh°IoDlThe (57)
p (l + klre)(l + Tth)(l +k2rh)D2

In a similar fashion, if we define

the thermal generation of excitons

assuming thermally generated holes
reaction scheme shown in Figure 37

photogenerated holes.

the rate constant of

as k2, we get:

(58)

also follow the same

as in the case of

158

It is known that interfacial pH is determined by the

equation:

pH (surface) = pH (bulk) + §%. (59)

where e is the electronic charge, C is zeta potential,

k is Boltzman constant and T is absolute temperature
(Hartley and Roe, 1940). Since zeta potential of BLM
(phospholipid) is found to be small, less than 5 mV
(Hanai, Haydon, and Taylor, 1965; Bonting and Bangham,
1967), the correction is unnecessary. Thus, the concen-
tration of hydrogen ions at the surface of BLM is

assumed to be proportional to the bulk concentration:

'H = ACO’H (60)

C dark, bulk

o
d
where A is the partition coefficient of protons at
hydrocarbon/water interfaces.

Finally, the substitution of Equations (54) and
(58) into Equation (25), and Equations (57) and (60) into
Equation (26), and then the resulting equations into

Equations (17) and (18) gives:

159

 

 

 

 

r
-€C l
Vtotal _ V = _ Gl (RT lpThEIOe )(1 _ e-GZt/Cm)
c O G1 + G2 F gt
0
r
r -8C 1
G2 RT C klkzteTthrhsIoe
+——-—G+G{F~ln OH }
I
l 2 (1 + klre)(1 + thl)(l + kZTh)D2ACd,bulk
-G t/C
x(l-e2 1m) (61)
for 0 < t < t
— — o
and
r
-eC 1
Vtotal V _ _ Gl (RT 1 ThEIoe )(l _ e-Glt /Cm)
c 0 G1 + G2 F kt
o
-Gl(t - tO)/Cm
x e
r
r —EC 1
G2 RT C klkZTeTthThCIOe
+G+G{§_ln 0H}
I
l 2 (l + klre)(l + thl)(l + szh)D2ACd,bulk
-G t /C —G (t — t )/C
x(l-e20m)e2 ° m (62)

for t i t.

160

The Effect of Temperature on the
Biphasic Photopotentials

 

 

The Effect of Temperature on
the Dark Membrane Potentials

 

 

For the purpose of analyzing the effect of tempera-
ture on the biphasic photopotentials using the theory
developed in the previous section, it is necessary to
understand the relationship between temperature and
protonic membrane conduction. In order to accomplish this
purpose, naturally, it is very helpful to establish the
effect of temperature on the mobility of protons, which
control both dark membrane potentials and mode of biphasic
photopotentials.

According to Equation (2), the dark membrane poten-
tials are the function of temperature, ionic transference
numbers, and ionic activities. If we assume the last two
terms are independent of temperature, the membrane
potentials (Em) should be linearly dependent on tempera-
ture (T). Shown in Figure 24, however, the membrane
potentials have the following relationship with
temperature:

- Ea/RT

Em(T) = Em'oe (63)

where Em o is a constant and Ea is the thermal coefficient
I

of the development of the dark membrane potential and R is

the gas constant. Thus, Equation (2) requires corrections

161

to account for temperature variations, indicating either
one or both of the following relationships may exist:

- E +/RT
t+(T) = tge t (64)

and

- E /RT

(T) aMc1

a _ (65)
MCl — aMCl,oe

The dissociation constant of HCl is very high and
its temperature dependence is fairly low. Although the
partition coefficient of protons at the edge of the
membrane may depend on temperature, the membrane and
aqueous phases are symmetric except for the HCl concen—
tration gradient. Thus, the substitution of Equation (65)
into Equation (2) for the inside and outside HCl
activities results in the cancellation of the thermal
effect on the partition coefficient if we assume the
effect of temperature on the partition coefficient is
constant for the dilute salt concentrations such as those
used in the present work. Furthermore, taking logarithm
of the Equation (65) as indicated by Equation (2)
eliminates the possible exponential relationship with
temperature as expressed in Equation (63), if activity
is temperature dependent. Thus, the possibility of the

contribution of the thermal effect on the HCl activity

162

and proton partition coefficient at the surfaces of BLM
can be safely eliminated.

The transference numbers of ionic species must
depend on the permeability through electric double

layers and the hydrocarbon layer or:

+ + +

t = f (td, th) (66)

where t; is the transference number through the electric
double layers and t; is that through the hydrocarbon layer.
As mentioned in the above paragraph, the partition
coefficients at the inner and outer surfaces of the
membrane can be assumed to be equally influenced by
temperature under the present experimental condition:
there is no electrical or adsorptive asymmetry across
the membrane except the HCl concentration gradient, within
the concentration range of 10-5 - 10.4 N. Thus, the effect
of temperature on the partition coefficient, hence, t3,
is negligible.

Equation (66) may now be written as:

+

+
t - f (th) — th

(67)

Now, we should recall that t+ is the transference
number of protons with respect to chloride ions. Thus,

in addition to Equation (67) we have:

t = f (th) = th (68)

163

for chloride ions, and also expressing the effect of

temperature we have:

- E +/RT
+ _ + t
th(T) — th,oe h (69)
and
_ _ - Et-/RT
th(T) = th,oe h (70)
+ .-
where th,o and th,o are constants.

In dilute salt solutions the effect of increasing
temperature, generally, is to bring the values of trans-
ference numbers from both ends of the extremes (l and 0)
to the value of midpoint or 0.5 (Taylor, 1938). If this
is the case in the present model membrane, the trans-
ference number should decrease from unity as temperature
is elevated beyond room temperature. On the contrary,
the membrane potential increases, hence, the "effective
transference number" increases beyond unity violating
the requirement that t+ + t- = 1. This anomaly may be
explained in the following way. Unlike the bulk salt
solutions, the effect of temperature is not to influence
the ionic selectivity. This is reasonable since the ionic
selectivity is mainly determined by the electric double
layers which do not contribute to the present transference

number, meaning the electric double layer is not influenced

164

by temperature so as to result in an increase in the
partition coefficient of chloride ions. Then, an increase
in the membrane potential must be due to an increase in
the "effective proton transference number" caused by an
increase in protonic mobility in the hydrocarbon layer.

This can be expressed by the relationship:

th = u /(u + u ) (71)

where u's are mobilities in the hydrocarbon layer.
Furthermore, there is a relationship between the
mobility and the diffusion coefficient of charged

particles as pointed out by Einstein in the form:
u = DF/RT (72)

where D is the diffusion coefficient and expressed in
the form:

- ED/RT
D = Doe (73)

where ED is the activation energy for diffusion. Thus,
if the above reasoning is correct, the activation energy
of the develOpment of the dark membrane potentials should
agree with the activation energy of the diffusion of
protons in the hydrocarbon layer or in bulk hydrocarbon
saturated with water. There is substantial evidence in

support of the above interpretation. The diffusion constant

of water in hydrocarbon medium has exponential temperature

165

dependence (Schatzberg, 1965). Indeed, the activation
energy of the diffusion of water in hydrocarbon is about
4 kcal/mole (Schatzberg, 1965) and the activation energy
of hydronium ions in water and ice is around 2 - 3
kcal/mole (Eigen and De Maeyer, 1958). The agreement
between these values and the value obtained in the
present work for the development of the dark membrane
potentials (5 kcal/mole) can be considered as remarkable,
taking into account the solid-like structure of the

hydrocarbon layer of BLM. Thus, we may conclude:

Ea = Eu z 5 kcal/mole (74)

The Effect of Temperature on
the Membrane CondfiCtivity

 

 

The concept that the effect of temperature is to
influence protonic mobility through water channels in the
hydrocarbon layer of BLM is also supported by the observa-
tions made on the temperature dependence of the dark
membrane conduction. The conductivity, 0, is given by
the sum of the contributions of the negative charge

carriers and positive charge carriers:
0 = Gl/S = n-eu- + n+eu+ (75)

- + . . .
where n and n are the den51t1es of the negat1ve and
positive charge carriers respectively, u- and u+ are the
respective mobilities, e is the magnitude of the electronic

charge, 1 is the thickness of the BLM, S is the area of

166

BLM, and G is the conductance. Due to the unity of the
proton transference number, Equation (75) can be

rewritten as:

O = n eu (76)

Temperature dependence of membrane conduction shows the
exponential temperature dependence given by:

- EO/RT
C(T) = Ooe (77)

where EC is an activation energy for conduction. Since
0 as given in Equation (75) is the product of the charge
density and the mobility, the exponential temperature

dependence of 0 could be due to the following two

possibilities:
- E /RT
n(T) = noe n (78)
and
- Eu/RT
u(T) = u e (79)

0

where En is the activation energy of charge generation.
Hence, the conductivity can be expressed as:

— (En + Eu)/RT

o(T) = noeuoe (80)

167

Since the experimental determination of Ea under the same

experimental condition (the constant n indicates that

H)
the Eu has about 5 kcal/mole, Equation (74). Since this
is in good agreement with the value of E0, we may

conclude:

E0 = Eu 2 7 kcal/mole (81)

Thus, the conductivity is determined by the charge
mobility, which, in turn, is determined by the state of
water channels in the hydrocarbon layer as mentioned in
the previous section, and which will be discussed in
detail shortly.

It is possible to evaluate permeability constant
(P) from the knowledge of the transference number (t)
and the bulk concentration of the ionic species (C) with

the following equation (Miyamoto, 1966):

P = tRTG/FZCS (82)

Since the specific conductance of the membrane for 10-1

° N HCl was 10.7 mhos/cmz,

M KCl and 10‘1 M NaCl and 4 x 10‘
and the transference numbers were 0.5 for the first two
salts and l for HCl at room temperature (23°C), from
Equation (82) we get:

_ _ _ -8
PK - PNa — PCl — 3 x 10 cm/sec (83)

168
and

PH = 2 x 10"4 cm/sec (84)

where P's stand for the permeability coefficients of
corresponding ions. Thus, the proton selectivity of
this model membrane is now well established.

The concentration of all-Epapgfretinal may be
assumed to be equal to bulk lipid concentration, which is
10-4 M/cm3. This value is reasonable in view of the
surface excess of the surfactants studied by Tien (1968c):

the value was of the order of 10"4

M/cm3, and retinal can
be considered as a surface active component in the present
model membrane. The maximum value of the hole concentra-
tion is not likely to be more than 10"7 M/cm3, since the
quantum efficiency of the most photosensitive organic
materials is less than 10'"3 charge carriers/photon.

Taking the limiting concentration of holes as 10-7 M/cm3,
and the transference number Of holes as unity, the con-
ductance of the same order of protonic conductance can
be obtained with the hole permeability coefficient:

Ph = 2 x 10-1 cm/sec (85)

Comparing Equation (85) with Equation (84) we get:

10 (86)

'UI'U
:EU‘
to

169

This is the main reason why it is plausible to assume
that the effect of temperature on the conductance of the
hole is negligible compared with that of protons: it is
known that the larger the permeability coefficient, the
smaller the thermal coefficient of conductance (Taylor,
1938). At this point, it is appropriate to state that

from Equation (81) we get:

E = Eu 2 7 kcal/mole (87)

and

E z 0 (88)

It may be asked whether high proton mobility is
due to the presence of water channels in the hydrocarbon
layer or not. In order to answer the question, first
let us suppose that there are enough water molecules in
the hydrocarbon layer so that the mechanism of proton
transfer is essentially similar to the diffusion of proton
in the bulk water; then, let us examine whether this
assumption can lead to the feasible explanation of the
experimental observations.

Since there is a close relationship between the
permeability coefficient and partition coefficient given

by:

P = uART/lF (89)

170

knowing proton mobility in water is 3 x 10-3 cm2/V sec
(Eigen, De Maeyer, 1958), and assuming that the partition
coefficient is unity and l is 45 A, we can evaluate the
permeability coefficient of proton in the water layer
with comparable thickness of the hydrocarbon layer of

BLM as:

PH = 4 x 102 cm/sec (90)

Comparing Equation (90) with Equation (84) it becomes

quite clear that the hydrocarbon layer of BLM is not

quite similar to the water layer. In order to obtain the
proton permeability coefficient of the present BLM, keeping
the same diffusion mechanism, the partition coefficient

must be less than unity or:

A = PH, BLM/PH, H20

5 x 10"7 (91)

If there is protonic transfer through water chains in the
BLM hydrocarbon layer, this partition coefficient also
should exist in the present BLM.

The partition coefficient can be defined as:

: h w _ - U/RT
A _ cH/cH — e (92)

where CE is the proton concentration in the hydrocarbon

layer, Cw

H is the proton concentration in water, and U is

171

the energy required for the proton transfer across the
interface. U can be estimated as the energy of the
transfer of a charge (q) as a function of the dielectric
constants of water (SW = 80) and hydrocarbon layer (eh),
taking the radius (r) of hydrated protons as 9 A by the

following equation (Gurney, 1962):

2
_5_ l.-l_
h w

At room temperature, T 2 300°K, the substitution of

Equation (91) into Equation (92) gives:
U 2 4 kcal/mole (94)

The resubstitution of Equation (94) into Equation (93)

gives:

8 z 3.6 (95)

The question now is whether Equation (95) is
acceptable or not.

The dielectric constant of the hydrocarbon layer
of BLM can be evaluated from the experimental value of
membrane capacitance (Cm) by the relation (Tien and
Diana, 1968):

EhS

Cm = _FI (96)

as

172

where S is the membrane area and l is the thickness of the
hydrocarbon layer. The value of the capacitance of all-

trans-retinal BLM was reported by Tien and Kobamoto (1969)

 

as 0.6 x 10_6 F/cmz. Taking this value and l = 45 A as

before, we obtain from Equation (96):

e z 3.5 (97)

A good agreement between Equation (95) and Equation (97)
may be taken as an indication of the validity of the
assumption made in deriving Equation (95), i.e. there may
exist water channels, through which proton transfer can
take place in almost crystalline structure of the hydro-

carbon layer of all—trans-retinal BLM.

 

The Effect of Temperature on
the Mode of Photoresponses

 

 

In the previous two sections it was concluded that
an increase in temperature increased the proton mobility;
thus leading to an increase in the membrane conductance,
mainly the G2 or protonic component. In this section we
will examine how this variation of membrane conductance can
modify the mode of biphasic photoresponses of all-trans-
retinal BLM.

First let us consider the extreme cases of R2 = 0
or R1 = 0. In these cases there were only monOphasic
photoresponses, and upon light-off the decay of the

potential started instantaneously. R2 = 0 was seen at

173

lower temperatures while R1 = 0 at higher temperatures.
These observations can be easily predicted from Equation

l>> G2 for the case of lower

temperature and G22» Gl for the case of higher temperature.

Then, the time course of the photoresponses must follow

(61) and (62) by putting G

an almost similar one to Vi - V0 for the lower tempera-
tures and v: - V0 for higher temperatures.

At the temperature range at which R1 and R2 can be
seen in a single recording, G1 and G2 are about the same
magnitude, indicating that both of the components, Vi - V0
and v: - V0, can impose an influence on the other com-
ponent. Now, let us examine how various aspects of R1 and
R2 can be predicted from this theory, and how these pre-

dictions fit with the observations made in this work.

From Equation (61) we get:

bt

AV = vC - v0 = - A(l - e‘at) + B(l - e" ) (98)
Then, dAV/dt = -aAe'at + bBe—bt (99)
At the minimum dAV/dt = 0, and Equation (99) becomes:

aAe—at = bBe-bt (100)

By solving Equation (100) for t and noting that t is now
the time required to reach the minimum of the photo-

potential we get:

174

= l
min b - a

 

(ln bB - ln aA) (101)
Since tmin must be greater than zero, we get:

bB > aA (102)

The substitution of Equations (19) - (22) into

Equation (l02) gives:

V - V
l o 1/2
G2 > Gl(v;—:—v;) (103)

Equation (103) gives the conditions at which the minimum

potential can be observed. These conditions are:

G2 > Gl 1f Vl - VO = V2 - V0 (104)
V2 - VO > Vl - Vo 1f G1 = G2 (105)
or G2 > G1 and V2 - VO > V1 - V0 (106)

The substitution of Equation (101) into Equation (61)

 

 

gives:
2
. G G G (V -V )
1n _ _ l _ _ _ l 2 2 o
VT - Vo _ G_:G—(Vl VoHl exp[ G -G ln ‘7—_—_—_ l}
1 2 2 1 G (V -v )
1 1 o
c; G G2(V -v )
2 2 2 2 o
+ 5—15—(V2-V0){1‘8Xp[- G 'G 1n ]}

 

2
1 2 2 1 G1(Vl-VO)

(107)

175

This potential was defined as the R1 component in the
present work by Equation (3) (see also Figure 39).
Equation (107) predicts that the R1 component decreases
with an increase in G2 exactly as observed in the present
experiments of the temperature dependence of the photo-
responses. It can also be seen that Equation (101)
predicts that the time required for reaching the minimum
potential (the peak time of R1) shortens with an increase
of G2. This point was repeatedly confirmed in the
present work.

Next, let us examine the R2 component and the post

effect. The post effect may be defined as the potential

development beyond the light-off time. From Equation (62)

we have:
AV = - ACe- a(t — to) + BDe- b(t - to) (108)
where
-G t /C
C=l-elom (109)
and
-G t /C
D = l — e 2 O m (110)
Then:
dAV/dt = aACe - bBDe (111)

At the maximum:

dAV/dt (112)

ll
0

176

Figure 39.--The analysis of the time course of the
biphasic photoresponses of all-trans-retinal

BLM. See the text for the details.

178
Substituting Equation (112) into Equation (111), we get:

—a(t - to) -b(t - to)
aACe = bBDe (113)
Solving the Equation (113) for t and designating
t = t , the time required for reaching the maximum

max

potential, we get:

t = t + __l——-(1n aAC - 1n bBD) (114)
max 0 a - b

Furthermore, by defining:

At = t - t (115)

we get:

At = a E b (ln aAC - 1n bBD) (116)

 

Since the post effect, by its definition, must occur at:
At > 0 (117)
Equation (116) becomes:

aAC > bBD (118)

The substitution of Equations (19) - (22), (109) and
(110) into Equation (118) gives:

-G t /C -G2t /C

2 l o m 2 o m
AV2 < Gl(l - e )AVl/G2(l e ) (119)

179

Equation (119) specifies the conditions at which the

post effect can be seen. These are:

AV2 < Vl 1f G1,: G2 (120)
G2 < Gl 1f Avl = Av2 (121)
or G2 < G1 and AV2 < Avl (122)

The substitution of Equations (19) - (22), (109), and

(110) into Equation (116) gives:

 

-G t /c
c G2(V —V)(1-e 1° m)
At = ———§——— ln 1 l O (123)
G1 ‘ G2 2 ’Gzto/Cm
G2(V2 -- VO)(1 -- e )

Noting the conditions (120) - (122) for the existence
of the post effect, the time required for reaching the
maximum potential after the light-off increases with
elongation of the illumination period, reaching the

maximum value given by:

 

2
c G (v - v )

At = §——¥—5— ln % l O (124)
1 2 G2(V2 - v0)

Substituting Equation (123) into Equation (62) and

subtracting Equation (61) after putting t = to' we get:

180

 

 

G -G t /C
l l o m
V - V = - (V - V )(1 - e )
max tO G1 + G2 1 o
-G t /C
G G2(V-V)(l-e 101“)
x[ex (- ———£——— 1n 1 l O )- l]
p G1 + G2 2 ’Gzto/Cm
G2(V2 - VO)(1 - e )
G -G t /C
2 2 o m
+ G_—:_G— (V2 - Vonl - e )
1 2
-G t /C
G G2(V-V)(l-elom)
x[ex (- -——3——— 1n 1 l O ) - l]
9 G1 + G2 2 -G2tO/Gm
G2(V2 - VO)(l - e )
(125)

From Equation (125) and the conditions given in Equations

(120) - (122) it can be said that V - V increases
max tO

with an increase in G2. This characteristic was clearly

demonstrated in Figures 22 and 23, where the post effect
increases with increasing temperature, which, in turn,
increases the value of G2. Not only At but also Vmax -

Vt increase with an increase in to. This point was
0

supported by the observation that keeping other experi-
mental parameters constant, the elongation of the illumina-
tion time alone resulted in the appearance of the R2
component, which was not observed during the shorter

time of illumination.

181

Finally, let us examine the temperature dependence
of R2/Rl. The substitution of Equation (124) for t - tO

of Equation (62) gives R2 or:

 

 

G -G t /C G
____1___ .. _ 10 m ___l___
R2 — G + G (Vl Vo)(l e ) exp ( G _ G 1n .
1 2 1 2 y m
T;
-G t .C 7
2 l o m i
cil—(v:L vo)(l e )) -:
-G t77C a
2 2 o m '
J;
G -G t /C
2 2 o m
+G + G (V2 ’ V1”l ' e )
1 2
-G t /C
G G2(V-V)(l-elom)
xeXp('G"Ԥ"cs-ln l l O -Gt/C)
l 2 G2(V _ V )(1 _ e 2 o m)
2 2 o
(126)

Dividing Equation (126) with R1 given by Equation (107),

and rearranging, we get:

182

R2/R1 = {GlAVl(l - e—GltO/Cm)} Gl/(GZ ' G1)
X {G2Av2(1 _ e-GZtO/Cm)}-Gl/(G2 — G1)
X {(Gl/G2)G2/(G2 - Gl)- (cl/62>Gl/(G2 - Gl)}/
[GlAv1_G2AV2—(GlAVl)G2/(G2-Gl)(G2AVé)-Gl/(G2-Gl)
X {‘Gl/G2)G2/(Gz -Gl)- (cl/G2>Gl/(G2 - G1)}1

(127)

Since biphasic responses must satisfy the conditions shown
in Equations (104) - (105) and (120) - (122), it is

permissible to assume:

GlAVl = G2AV2 (128)

The substitution of Equation (128) into Equation (127)

gives:

‘G t /C (G /G *1) -G t /C (G /G -1)
- R2/Rl = (l-e l O m) l 2 (l-e 2 o m) 2 l

(129)

Now, it is possible to predict the variation of the

ratio as a function of changes in G2.

14”. .4,__.,.__....‘.A~-.-1 ‘ ....-_
.
I

av-..

183

When G = G

2 1’ we get:

- R2/R1 = 1 (130)

At this point we introduce the factor of temperature.
It was found in the present work that the following

relationship can hold:

- E /R'I‘
G2(T) = Gge G2 (131)

o . . . .
where G 18 constant and E is the act1vat1on energy for

2 G
2
protonic conduction. Assuming that G1 is temperature
insensitive or that E ‘« EG , the substitution of

G
l 2
Equation (131) into Equation (129) gives:

 

G1 eEGZ/RT
-Glto/Cm (—3 _ l)
- R2/Rl = (1 - e ) G2
0 o
-tOG2 -EG /RT G2 -EG /RT
C e 2 (a— e 2 ’1)
x (l - e m ) l
(132)

As T increases the equation (132) behaves differently,

depending on the relative values of G1 and G3. When

> G : (133)

G1 2

- R2/R1=$ 0 (134)

A. 1' - ‘0‘ 'g .‘a-

.f-yL‘I I... ‘

184

0
When G2 > G1. (135)
-E /RT -E /RT
-GltO/Cm -1 -(G3tO/Cm)e G2 G: e G2
-R2/R1 = (l-e ) (l-e )(—— -1)

G
1

(136)

Equation (134) does not agree with the observations made
in the present work, while Equation (136) shows rapidly

increasing function with respect to an increase in

 

temperature, like the curve shown in Figure 21. Thus,
it can be concluded that at the condition given by
Equation (135), i.e., relatively higher protonic conduc-
tion than hole conduction, the theory develOped in this
work predicts rather satisfactorily. Since the generation
of the R2 phase required higher protonic conductance, the
requirement given by the Equation (135) should be con-
sidered as a reasonable prediction agreeing with
experimental observations.

It is clear that the theory developed in the
present work predicts and explains all the observations
made as a function of temperature.

The Thermal Coefficients
of’Fhotoresponses

The thermal coefficient of R1 was far smaller than
that of R2 and in the presence of sufficient electron

acceptors was reduced to almost zero. If R1 would show

185

temperature dependence, it was always a negative type:

Rl decreased with increasing temperature. Because of the
high hole permeability coefficient as shown in Equation
(86), it can be assumed that the hole potential AVl is
almost temperature independent. Thus, the negative
temperature dependence of R1 may mean either the presence
of the essential effect of temperature on the generation
of holes, or the effect of increasing protonic potential
or AVZ. The first possibility is not likely to be

feasible with negative temperature dependence. Rather,

[an-nuap—‘J-u— »_..¢--.- _ I‘m-‘w. -_~.— -—
.. - ‘ ., V -. -

since the photoresponses, R1 and R2, can be seen at the
temperature above 40°C under proper experimental condi-
tions, and the magnitude of this negative thermal
coefficient was almost equal to that of the protonic
resistance change as shown in Table 4, it is more likely
that the second possibility is the case in the present
model BLM. Thus, the analysis of thermal coefficient of
R2 will provide the information valuable in understanding
that of R1: R2 is mainly determined by the temperature
dependence of the protonic potential.

The thermal coefficient of R2 was larger than both
that of R1 and of the dark membrane potential development,
and hence, also larger than that of the membrane con-
ductance as shown in Table 4. Our task here is to under-
stand why this is so. Examining Equation (126) shows
that in order to get a single thermal coefficient we

must have:

186

G2(T)

R2 (T) a Gl(T) + G2(T) (137)

 

and other terms of Equation (126) must be negligible.

Let E <« E (138)
G1 G2

 

and G1(T) = G1 (139)
then, since at elevated temperatures:
2 2

G2(T) »>Gl (140)

we get:
G1
From Equation (81) we have:
2 *
F uH(T)CH(T)S
G2(T) = l (142)
. * m _ _ -U/RT _ -U/RT

Since CH(1) — ACH - CHe — CH(T)e (143)

where CH(T) is the concentration of the protons

including the photogenerated protons, and U is the energy
required for the transfer of proton across the water/
hydrocarbon interface. Thus, Equation (142) should have
three exponentially temperature dependent terms which may

appear as a single temperature coefficient as:

YET—5-4—u— -‘unta -—---..... aA-bmn

187

ER2 = Eu + U + EC (144)
where Eu is the thermal coefficient of proton mobility

in the hydrocarbon layer and EC is the thermal

H
coefficient of the proton generation by the oxidation
of water molecules. Since ER2 z 16 kcal/mole, the

substitution of Equations (74) and (94) into Equation

(144) gives:

EC z 7 kcal/mole (145)
H

There are two points of interest:

1. Why cannot U be recognized in the thermal
coefficient of the development of membrane
potential and that of dark membrane
conductance?

2. What is the significance of the value of
ECH given in Equation (145)?

The first problem can best be understood by considering
the effect of buffering. The dark parameters were
measured at steady states at which it is possible to
have equilibrium between the transfer of proton into the

hydrocarbon layer and the proton transfer into the bulk

aqueous solution by buffering capacity:

ng/dt - dCE/dt = o (146)

_.‘_..L._._7____.r.-A_.. -‘. _.‘—’.'

188

at the dark state. It is reasonable to assume that this
equilibrium can exist at all temperatures if we allow the
system to stay at a stable state. Hence, the positive
effect of temperature on the proton transfer into hydro-
carbon can be unobservable: this does not necessarily

require:

U = E (147)

where Eb is the thermal coefficient for buffering. On
the other hand, at illumination where the rate of the
generation of protons is greater than the rate of

buffering into the bulk aqueous solution, instead of

Equation (146), we have:
H H
dCU/dt - dCb/dt > O (148)

at the illuminated state. Under this condition it is

obvious:

U > E (149)

since the former process requires the transfer of
charged species across an interface, while the latter
process is simple diffusion in the same aqueous phase.
Thus, the thermal coefficient of the partition of ions
can become observable in the case of photoresponse. The

above discussion is feasible, since the buffer capacity

189

closely controls generation AV and, hence, R2 as

2

discussed in a previous section. The second question may
be answered in the following manner. Since the hydrogen
bond of a water—carbonyl group is about 3 kcal/mole

(Karyakin, Muradova, Sayenko, 1969), EC can be considered
H

as a reasonable estimation of the activation energy of

I“

proton generation. This lower oxidation energy may be

. rm..-‘ u . as,-~1

due to the fact that the interface in the presence of

excited retinal can function as a catalytic site for the

‘A-.- “It: '0 n:

q.”
1

generation of protons and possibly oxygen.

The energetics of generation of the charge carriers
(holes and protons), discussed above, are schematically
shown in Figure 40. Photoexcitation is considered to
occur only via ground state singlet-lst excited singlet
excitation, then intersystem transfer to the lst excited
triplet occurs. This is due to the fact that since there
is good agreement between the action spectrum of photo-
voltaic and photo conductive effects of the all-trans-
retinal BLM with the absorption spectrum of bulk lipid
solution (Tien and Verma, 1970), we may conclude that the
ground state singlet-lst excited singlet is the major, if
not the only, excitation pathway, and the ground state
singlet-lst excited triplet transition may not exist in a
significant probability. Secondly, this is due to the fact
that the triplet state is not directly involved in the

process of charge generation, as can be seen in Figure

ii".

A

L

I
rq‘D.¢t~'\o;.“(.¢. are, l...~1li,.no

190

Figure 40.--The electronic energy levels of all-trans-
retinal bimolecular lipid membrane. The
values of lst excited triplet was from

Rosenberg and Harder (1967).

A - electron acceptors (ferric ions)
D - electron donors (water molecules)
L - hydrocarbon layer

r - all-trans—retinal molecules

SO - ground state singlet

Sl - lst excited singlet

T - lst excited triplet

F ‘1,

. . . . u
'¥.|4(IJJ.)I«II 5.. (0:7)!an (5|)

 

L

 

 

 

 

. _ _

 

_
2 1
3 30v m_o>on >otocm

0

Reaction Coordinate

192

40. Triplet excitons are trapped in the surface sites

as charge transfer complexes. It was assumed in deriving
the kinetic equation that the concentration of this com-
plex and triplet excitons can exist in equilibrium due

to the trap depth of approximately kT. The charge
generation step is the charge separation of this complex
which can be activated by the thermal energy and not the
direct generation of the lst excited singlet or lst

excited triplet, since even the activation energy of semi-

'_-. f1“... .. ..._._._ .._ .__- l T..._- -_——'

conduction of the present BLM was lower than the energy
levels involved in these two processes. The thermal
coefficient of R1 is always less than that of R2; in the
presence of high Fe3+ concentrations zero activation
energy was observed. Thus, not only the thermal coeffi—
cient of hole diffusion but also that of hole generation
is so low as to be insignificant. At lower ferric ion
concentrations the thermal coefficient of R1 showed the
negative of that of R2, and at higher ferric ion
concentrations, the thermal coefficient was almost zero.
This may be interpreted to mean that the negative
activation energy is due to the negative contribution of
the R2 component. The degree of their contribution may
depend on the relative concentration of the two charge
carriers. That is, if ferric ion concentration is low,
the R2 effect is greater while at a higher ferric con-

centration the effect of R2 is negligible. From this it

193

is possible to regard that the activation energy of true
R1 is far smaller than that of R2, or less than kT

(10-2 ev). After charge generation, the triplet

excited state, which has approximately the same energy
level as the triplet state, diffuses across the hydro-
carbon layer. Upon reaching the outer surface, the
thermal activation of about 0.3 eV leads to generation

of protons. The energy levels of the lst excited singlet

and lst excited triplet may be considered the same as

those in crystalline all-trans—retinals or 3.4 ev and

 

2.0 ev, respectively (Rosenberg and Harder, 1967).
However, this triplet state has a considerably higher
value than that of an all-Erans-retinal solution in
non—hydroxyl solvent, which is 0.8 ev (Abrahamson and
Ostroy, 1967).

The Further Experimental
Verification of the Theory

 

The experimental verification of the theory can be
seen in the following four points, in addition to the
various aspects of temperature dependence discussed in
the previous section.

The Effect of Light Intensity:
Logarithmic Dependence

 

 

In order to find the relationship between the R1
and R2 components and light intensity, first, let us

obtain R1 by rearranging Equation (93):

194

R1 = EE—x—EE [G2(V2 - v0) - G1(Vl — v0)
-G /(G - G ) G /(G - G )
1 2 1 2 2 1
+ {G2(V2 - vo)} {Gl(V2 - vo)}
G /(G - G ) G /(G - G )
{(Gl/Gz) l 2 l - (Gl/GZ) 2 2 l }] (150)

and R2 by rearranging Equation (126):

_ G /(G — G )
———i———{G (V _ V )(1 _ e Glto/Cm)} 2 2 1
G1 + G2 1 l 0

R2

_ -G /(G - G )
GZtO/Cm 1 2 l

{G (v2 - v0)(1 - e )}

G /(G - G ) G /(G - G )
{(Gl/GZ) 2 2 l - (cl/62) l 2 l } (151)

From Equation (150) we obtain:

195

de/dlnIO = {RT/F(G + G2)}[G — G

1 2 1
G /(G — G > G /(G - G )
1 2 1 2 2 1
— G /(G - G ) G /(G - G )
1 2 1 2 2 1
G2 G1
-G /(G -G ) G /(G -G )
1 2 1 1 2 1
x{(V2-VO) {G2/(G2-Gl)}(Vl-VO)
G /(G -G ) -G /(G -G )
2 2 1 2 2 1
- (Vl-VO) Gl/(GZ-Gl)(V2-VO) }]
(152)
If we assume (Vl - V0) = (V2 - V0), we get:

62/(62 - G1) -Gl/(G2 - G1)

de/dlnIO = {RT/F(G + G2)}Gl G

l

G /(G - G ) G /(G - G )
1 2 1 2 2 1

(153)

Similarly from Equation (151) we get:

 

*““fi

-473-‘9'

196

.1327.

dR2/d1nIO F

{Gl(1

_ G2to/Cm

{G (l - e )

2

Gz/(GZ ‘ Gl

{(Gl/GZ)

-G /(G -G )
1 2 1
[(VZ—Vo)

Gz/(G2_Gl)

-(vl—VO) {

V

2

If we assume Vl - VO

dR2/d1nIO {RT/F(Gl + G2)}{Gl

- G t /C
{G2(l — e 2 O

G /(G
2 2
{(Gl/GZ)

-e

V
O

m)}

- G

G /(G - G )
- GltO/Cm 2 2 1
)}
- Gl/(G2 - G1)
}
'7‘“)
) G /(G - G ) i
1 2 1 !
- (Gl/GZ) } g
i
G /(G -G )
1 2 1 . .
{GZ/(Gz-Gl)}(Vl-VO) cw
-G /(G —G )
2 2 1
Gl/(GZ—Gl)}(V2-VO) 1
(154)
, we get:
_ G /(G - G )
GltO/Cm 2 2 1

(l-e )}

 

)

G /(G -G )
1 2 1
— (Gl/GZ) }

1

(155)

197

The Equations (153) and (155) indicate the lepe of the
straight line which can be obtained by plotting either R1
or R2 against the logarithm of light intensity. It is
shown in Figure 19 that such a linear relationship exists,
suggesting the validity of the theory.

If we define:

dR2/dlnIO
de/dlnIO

 

d = (156)

then the substitution of Equations (153) and (155) into

Equation (156) gives:

G /(G - G )
d = {1 - exp(-GltO/Cm)} 2 2 l

-G2/(G2 - G1)
x {l - exp(-GZtO/Cm)} (157)

Solving for G1’ we get:

_ -1
G1 — (ZtO/lend + 1/G2) (158)

At the conditions specified in the legend, Figure 19

gives:

a = 1.5 (159)

198

with the substitution of Equation (159) into Equation

(158), noting that tO is one second, Cm is 0.6 uF/sz,

and G2 is 5 x 10-8 mhos cm_2, we get:

G1 = 3.7 x 10‘8 mhos cm‘2 (160)

- '.' m,

This prediction is reasonable since the steeper slope of
R2 must arise from slightly greater G2 with respect to
G1' From the above discussion it is clear that the light

intensity dependence of photo-responses can be

“T'""’“ n. ..

satisfactorily explained by the present theory.

The Effect of the Outside pH

 

From Equations (150) and (151) we get:
O——
de/dlnCH — {RT/F(Gl + G2)}

G Gl/GZ - G1) GZ/(Gz - G1)}

1
[G2 - G2 _ Gl{(Gl/G2) - (Gl/Gz)

G /(G - G ) -G /(G - G )
2 2 1 2 2 1
x {Gl(Vl - vo)} {G2(V2 - vo)} ]

(161)

and

199

G /(G -G )
o -G1to/Cm 2 2 l
dR2/dlnCH = {RT/F(Gl+G2)}{Gl(Vl-VO)(l-e }
G /(G - G ) G /(G — G )
2 2 l 1 2 l
{(Gl/Gz) - (Gl/Gz) }
-G t /C -G /(G - G )
{G2(l-e20m)} l 2 l
-G /(G - G )
2 2 1
[{Gl/(G2 - Gl)}(V2 - v0) ]
(162)
If we put Vl - VO = V2 - V0, the Equations (161) and
(162) become:
de/dlnCE = - RT/F(Gl + G2) [G2 - Gl/(Gl + G2)
G /(G - G ) G /(G -G )
l 2 1 2 2 l
G /(G — G ) - G /(G G )1
G 2 2 l G 2 2 1
l 2
(163)

and

200

/(G - G )
Mlt /Cm Gz 2 1
dR2/d1nCH={RT/F(Gl+ G 21)}{G (1 - e mfl
G /(G -G ) G /(G -G )
2 2 1 1 2 1
{(Gl/GZ) -(Gl/G2) }{Gl/(G2-Gl)}
- GZtO/Cm ’ Gz/(Gz ‘ Gl)
{ G2(l - e )}
(164)

This linear relationship between the photoresponses and the
logarithm of outside pH is clearly demonstrated in Figure
'29: an increase in the outside pH induces an increase in
R2 and a decrease in R1, both in a linear relationship.
Thus, this point again shows good agreement between

theory and the experimental observations.

The Effect of Buffer Capacity

 

Buffer capacity influences the photoresponses
dramatically. This effect may be best understood by
defining a coupling coefficient, 9, between the rate of
photogeneration of protons at the outer surface to the
accumulation at the inner surface. By referring to

Figure 36 and neglecting D4 and k5 as before we get:

9 = (02 - k8 - k5 - k9)/k2 (165)

201

If the buffer capacities of two aqueous phases are the
same, as they were in the present work, k8 = k9 and

Equation (165) becomes:

9 = (D2 - 2k8 - k3)/k2 (166)

Hence, the greater the buffer capacity, the smaller the
coupling coefficient, meaning that R2 decreases in spite
of the occurrence of the equally developed Rl component. ;

The Effect of Light Intensity: )
Threshold—Requirement

 

 

Pure Rl responses could be transformed to biphasic
responses by simply increasing light intensity without
changing other experimental parameters (Figure 19).

Can this fact be explained with the theory presented in
this work?

There are two possibilities for the explanation of
the requirement of threshold intensity for generation of
the R2 component. One is threshold requirement to
accumulate the photoproduct, or protons in the present
case, and the other is the threshold requirement for a
completely independent secondary process. The latter
possibility can be eliminated by the following considera-
tion. The only secondary process possible in the present
system is photoinduced proton diffusion. However, the
magnitude of R2 declined, after reaching a maximum under

the effect of increasing temperatures. If this component

202

is the proton diffusion potential, independent from the

initial hole potential, such a decline should not be

observed since the protonic conductivity increased

monotonously with an increase in temperature. The

presence of a maximum in R2 can be satisfactorily

explained if we consider that R2 is the resulting 2H2
potential of the initial hole potential. That is, an

increase in protonic conduction, which requires an 1
increase in hydration of the hydrocarbon layer, acts
unfavorably for the develOpment of R1 by decreasing the w
lifetime of holes in the hydrocarbon layer. This

reasoning leads to the concept that there must be a

threshold intensity for the initiation of the two com-

ponents and that R2 must have a higher threshold

intensity, since the accumulation of a certain concentra-

tion of photOproduct, protons, may be required for the

generation of R2. There is another consequence of this

reasoning: the maximum photopotential can be observed

at a proper protonic conductivity where the positive

effect of proton conductivity equals the negative effect

on the hole conductivity. Thus, the presence of

threshold intensity for the generation of R2 is in

support of the present theory, indicating that R2 is the

resulting diffusion potential of the accumulated protons,

generated by the holes which have diffused through the

membrane, which in turn have generated R1.

203

"Hole—Proton Diffusion" Mechanism
and Its Alternatives

 

 

The formulation of the present theory was based on
the fact that a pair of calomel electrodes can detect
only the potentials developed by the diffusion of charge
carriers. In order to use this concept the transference
number of protons was experimentally determined as unity
to chloride ions and the transference number of holes
was also assumed to be unity. The assumption made con—
cerning the hole transference number can be well justified
by the observation that the presence of a strong electron
acceptor (ferric ions) can have a definite effect on
determining not only the direction but also the magnitude
of R1, without exception, and agrees with the sign of
the potentials predicted by the assumption of the hole
diffusion from the side onto which ferric ions had
adsorbed. This explanation is equivalent to the concept
of hole injection into the hydrocarbon layer at the side
of ferric ions after the excitation of retinal incor-
porated into the hydrocarbon layer. Such a possibility
was also proposed to occur at semiconductor/water inter—
faces (Kallmann and POpe, 1960). The present work does
not prove the presence of electronic (electron and hole)
conduction in the all-trans-retinal BLM. Nevertheless,
the present work strongly favors the presence of hole
conduction by establishing the presence of protonic

potential, which is free from the effect of other ionic

204

species, and by demonstrating the presence of the non-
protonic potential, which was proposed to be the hole
potential. The fact that the non-protonic potential

could not be leakage potential of other anions or cations
was clearly demonstrated by the observation that the ten-
fold concentration gradient of KCl and NaCl did not have
any effect on the R1 and R2 components. Thus, the success
of the present theory in eXplaining all observed facts
indicates strongly that the assumption of the presence of
the hole conduction in all—trans—retinal BLM is well

justified.

CHAPTER VII

TOWARD A FURTHER UNDERSTANDING OF

EARLY RECEPTOR POTENTIALS

Most of the observations made in the present work
strongly suggest the basic similarities between the
photoresponses of all-transfretinal BLM and the early
receptor potentials of biological photoreceptors. In
this chapter, these relevant observations are fully
discussed, by referring to biological observations, in
relation to the mechanism of the generation of the early
receptor potentials and its relationship to the generation
of electroretinograms.

The basic idea to be examined is that the mechanism
proposed in this work for the explanation of the biphasic
photoresponses of all-trans-retinal BLM may have a
biological role as the primary process leading to genera-
tion of the receptor potentials. The proposed mechanism
may be summarized in short: the photoinduced redox
reactions generate, by the diffusion of holes, the fast
phase of the biphasic early receptor potentials, and the
transport of the photogenerated protons is responsible for
the slow phase of the potentials. The proton transport
may further induce ionic transport as a sequentially

205

206

coupled transport phenomena. This mechanism may be

called a redox-protonic-ionic excitation mechanism.

The Membranous Origin of the
Early Receptor Potentials

 

 

This investigation was undertaken in order to
examine the possibility as to whether the photoresponses
similar to the early receptor potentials could be generated
with the model membrane system like all-trans-retinal BLM.
It is remarkable that the observations made in this work
showed the striking similarities between the photo-

responses of all-trans—retinal BLM and the early receptor

 

potentials. This fact indicates that accessory structures
may not be necessary for the production of the early
receptor potentials and, hence, that the pigmented
membrane of rods and cones can generate the early

receptor potentials. The mechanism of the generation of
the early receptor potentials of the pigment epithelium
and in the plant cells may be different from that of the
rods and cones to a certain degree.

It is well established that the wave form of the
early receptor potentials is dependent on the amount and
kinds of rhodopsin and its derivatives in the rods and
cones (Cone, 1967; Pak and Boez, 1967; Hagins and
McGaughy, 1967). The point of special interest is that
the regeneration of rhodopsin derivatives can reverse the

wave form completely. The result was explained as the

“H
I

np _ . n .. —__‘.-_u‘..-_ 4....--1-
. « ‘ b ’

12.
l

.._...‘- “‘2‘
a.
,

207

recovering of the original charge distribution in and
around the pigment molecules. This point may lessen the
possibility of the generation of photochemical products
such as protons and may lead to the conclusion that the
early receptor potentials do not satisfy the condition
of the visual receptor excitation, since rhodopsin alone,
and not the derivatives of rhodopsin, determines the
excitation (Pak, 1968).

On the other hand, it is possible to introduce the
active role of the membranous structure into the inter-
pretation of the above observation. The generation of
rhodopsin derivatives necessarily produces changes in the
membrane structure so that the membrane surfaces are

changed from the original "rhodopsin state,’ causing
proton generation and transport to act in the reverse
direction. The structural changes of rhodopsin and its
derivatives would change the charge distribution at two
surfaces in a drastic manner. However, the salt con-
centration across the disc membrane is constant, regard—
less of the changes of the surface charge distribution

due to the formation of the rhodopsin derivatives. If

so, the generation of protons by the rhodopsin derivatives
may not act as efficiently as the rhodopsin itself since
the salt concentration gradient must be in an ideal con—
dition in the latter case. Thus, although the generation

of the early receptor potentials can be perfect, since

protons can be generated and transported to the same

208

extent as in the case of "rhodopsin state,‘ the following
cation transport may not be induced if the salt con-
centration gradient is not at the proper conditions in
view of the energy coupling or the characteristics of

the pores.

Lettvin (1965) prOposed that the shift of dipoles
with respect to the retinal surface or disc membrane could
serve as the mechanism of the generation of the early
receptor potentials. However, there seems to be no
supporting evidence for this type of molecular arrangement
of the pigment in the retinal disc membranes (Brindley
and Gardner-Medwin, 1966). Furthermore, in the pigment
epithelium and plant cell the pigment granules are in
random orientation with respect to the surface of the
tissues and the cell membranes (Figure 1). It is very
important to note that in the cell wall of the pigment
epithelium and the plant cell, pigment concentrations are
usually low. It is at this point that the rhodopsin
derivative hypothesis and dipole hypothesis encounter
immense difficulty: the "excitation" has to be transferred
to the non—pigmented cell wall from the pigment granules,
since the simple charge distribution along the disc
membrane due to the intramolecular charge redistribution
is not applicable to the other pigmented tissues. In the
mechanism develOped in the present work it was proposed

that photogenerated protons could perform the function

209

of the mediator of the early receptor potentials. Thus,
there is no difficulty of this type.

The possibility that electronic conduction along a
conjugated carotenoid chain could be involved in visual
excitation was first suggested by Dartnall (1948). The
mechanism of such an electronic process was speculated as
solid state photoconduction in photorecptor lamellar
structures by Chynoweth and Schneider (1954) and Reid
(1957). The first evidence of the existence of photo—
conduction in beta-carotene glasses was reported by
Rosenberg (1958). The same worker has established the
existence of photoconduction in the glasses of retinals
(Rosenberg and Harder, 1967) and dried rods (Rosenberg,
Orlando, and Orlando, 1961). Brown and Murakami (1964a)
and Crawford, Gage, and Brown (1967) suggested that
photoconductivity of the pigmented tissue could be the
mechanism of the generation of early receptor potentials.
Recently, Rosenberg and Postow (1970) pointed out a
similarity between the biphasic mode of the early
receptor potentials and the photoresponses of beta-
carotene glasses, and suggested that the two phases could
be due to the photovoltaic and photoconductivity effects.
This analysis is similar to a certain degree to that of
the present work, although the charge carriers in the
former case are electronic for both phases, while those
in the latter case are electronic for the first phase

and protonic for the second phase.

210

The Coupling of Electronic Transport
with Ionib Transport in
Visual Excitation

 

 

 

The possibility of coupling electron movement across
the biological membrane to ionic movement was first
proposed by Lillie (1915, 1916) in way of making an
analogy of the hypothetical metal electrode reversible to
the ionic species which would be responsible for generation
of membrane potential. About one decade later Lund (1928)
extended the theory to the development of a continuous
membrane potential with the mechanism of electron movement
through respiratory enzymes. Along with the support of
Lund's theory (Marsh, 1935), an ionic coupling theory was
developed to account for the accumulation of salts across
the biological membranes (Stiehler and Flexner, 1938;
Lundegardh, 1939; Ussing, 1947; Leve and Ussing, 1948).

In the 1940's Lund's theory was vigorously supported by
many workers such as Lundegardh (1945; 1946), Robertson
and Wilkins (1948a, b), and Crane, Davies and Longmuir
(1948a, b). Quantitative approaches to the Lund theory
have also been carried out by Conway and Davis and Ogston
(1950). The coupling of electron transport to ionic
transport and, then, to mitochondrial and photosynthetic
phosphorylation was proposed by Mitchell (1961, 1966).

It was Jahn (1962) who included non-enzymic electron
pathways (such as beta-carotene and vitamin A derivatives)

which could be coupled to ion transport. Further, Jahn

211

(1963) extended the possibility of the similar mechanism
to visual excitation. Jahn's (1962, 1963) theory of
visual excitation may be summarized as follows: light
absorption leads to the conformation change of retinal
from ll-gis to all-trans, thereby increasing the
probability of the transport of the electrons made
available through redox systems which could exist across
the membrane.

However, since the discovery of the early receptor
potential of the photoreceptors, the application of Jahn's
theory to the interpretation of generation of the poten-
tials has not been attempted. Judging from the observa-
tions made in the present work, it is more than likely
that Jahn's theory can be applied to the explanation of
the generation of the early receptor potentials. Further-
more, a three-step excitation mechanism may, indeed,
function in visual photoreceptors if electronic movement
can initiate protonic movement, which, in turn, generates
potassium and sodium movement as proposed in the present
work. Therefore, our next question is whether or not
proton can be generated by light in photoreceptors. This
point will be discussed in the following section. Section
D discusses the possibility of coupling between the

protonic transport and the sodium and potassium transport.

212

The Photogeneration of Protons
in Photoreceptors

 

 

There is ample evidence to indicate changes in the
proton concentrations of the aqueous media of the retina,
the suspensions of the outer segment of the photoreceptors,
and rhodopsin solutions upon illumination.

Frog retina, both in vitro and in vivo, became more
acid when exposed to light (Kobakova, 1946). A rapid fall
of pH occurred at the first 5-10 minutes of illumination
falling from pH 7.8 to 7.7 reversibly. Ostrovskii,
Fedorovich, and Polyak (1968) could not observe a
significant pH change after illumination of the frog
retina at the initial pH of 6.3-6.4, while the suspension
of the outer segments of the photoreceptor showed an
increase in pH ranging from 0.06 - 0.10 after a 3 minute
illumination at the initial pH of about 6.15. The
discrepancy between the observation of Kobakova (1946) and
that of Ostrovskii, Fedorovich, and Polyak (1968) may be
due to the effect of the initial pH of the bathing solutions
since it is known that this factor influences the mode of
the pH changes of rhodopsin solutions significantly, as
mentioned below. Nevertheless, it is evident that the pH
of the retina can be changed by exposure to light.

Further understanding of this aspect may occur if we know
the internal pH of the retina and the relationship between

pH change and the photoresponses.

213

pH of a rhodopsin solution could be also changed
by illumination. Radding and Wald (l955-l956a, b)
observed that the illumination with flashes of intense
light induced not only slow acid-base changes in rhodopsin
solution, but also an immediate rise in pH. This
immediate rise in pH was observed between pH 2 and pH 8
with a decrease in the magnitude at the neutral pH, and
was followed by the slow changes which were always in the
direction of neutrality. Studying the pH range between 4
and 10 Hara (1958) and Hara and Hara (1960) observed
similar results reported by the above workers. They also
observed that at an alkaline pH the immediate change of
pH was toward the neutral region. The slower component
stayed at a constant value of pH following an immediate
rise at the acid pH; on the other hand, it decreased
continuously toward the neutral pH at the alkaline region.
Hara (1958) also observed an immediate increase in the
resistance of the rhodopsin solution after exposure to the
flashes of intense light at alkaline pH as well as acid
pH. Recently, Fujishita (1968, 1969) observed the various
modes of changes in the pH of rhodopsin solution. With
dim green light a pH change could be induced without
bleaching rhodopsin. The direction of a pH change is
dependent on the temperature and the initial pH of the
solution. Fujishita defined an "isoionic point,"
decreasing with increasing temperatures. The values were

somewhat higher than those of the isoelectric point of

214

the frog rhodOpsin (pH 4.47) and those of the bleached
product (4.57), determined by Broda and Victor (1940).
McConnelL Rafferty, and Dilley (1968) reported that
proton uptake and extrusion by bovine retinal outer
segment fragments required the presence of unbleached
rhodOpsin while the stoichiometric ratio between proton
uptake and rhodopsin bleached was a function of experi-

mental conditions.

“in. “‘ .14-;__-.~..p -1
I

Most of the workers mentioned above tend to con- 3
sider the mechanism of pH changes as the exposure of the §_.;
acid—base groups to the aqueous phase due to the removal
of retinal and/or the conformation changes of opsin. Due
to the complex mode of the changes in pH on illumination,
however, further studies are needed to confirm this
interpretation. Nevertheless, an important question at
the present time is whether, regardless of the mechanism
of generation of photoinduced pH changes, pH can influence
the mode of generation of the early receptor potentials,
since if protons can function as the mediator there may
be an interaction between protons and the mode of the
photopotential generation via product inhibition.

The effect of pH on the early receptor potential
was first studied by Brindley and Gardner-Medwin (1966).
Thirty minute soaking of frog retinas in 120 mM phosphate
buffers between pH 8.5 and 2.8 at 24°C showed a rather
small effect on the mode of the early receptor potential.

The biphasic response could be observed over this pH

215

range. The magnitude of the two phases decreased and the
time required for the development of R2 became longer as
pH decreased. Pak, Rozzi, and Ebrey (1967), using the
frog retina and a 50 mM buffer system between pH 3 and

pH 11, observed, in general, a similar pH effect to that
noted by Brindley and Gardner-Medwin (1966). Furthermore,
Pak, Rozzi, and Ebrey (1967) observed that the amplitude
of R2 had its Optimum value near the neutral pH and
reversibly decreased at high and low values of pH. The
reversibility of responses decreased at the extreme ends
of pH's in such a manner that the early receptor potential
could be abolished after a 30 minute immersion. At the
higher pH the dominant component was R2 while at the lower
pH the response was dominated by R1. From these
observations they concluded that lowering pH not only
reduced R2 but also enhanced R1. Indeed, when the sample
was cooled to 2°C to eliminate R2, R1 was extremely
enhanced and prolonged at a pH lower than 4. Using the
eyes of albino rats and a 110 mM sodium phosphate buffer,
Arden, Bridges, Ikeda, and Siegel (1968) observed that
the variation of pH between 4.4 and 9.0 had little effect
on the early receptor potential. With the pigment
epithelium of the frog, Crawford, Gage, and Brown (1967)
observed a decrease in R2 and an increase in R1 as pH was
lowered. From these studies it may be concluded that pH
of the bathing solution has definite effects on the mode

of generation of the early receptor potential.

216

At the higher pH R2 tends to increase and at the
lower pH R1 tends to increase. At the neutral pH the
early receptor potential is optimum while the pH devia-
tion to both ends decreases the magnitude of the potential.

It was determined that proton is the medium of
generation of the early receptor potential because of
the following reasons:

1. Only H+, OH_, and H3O+ ions influence the

early receptor potential.

2. Protons can modify the wave form and magnitude
of the potentials drastically.

3. If indeed the transition from metarhodOpsin I
to metarhosopsin II is responsible for R2
generation, then, the ion generated in the
transition is proton (Cone, 1967).

4. Falk and Flat (1966) reached the conclusion,
after studying the conductance change of retina
and rhodopsin solutions on illumination, that
the faster component of the photoresponses in
their system was possibly protonic conduction.

In summary, it is very likely that the finding,
made in the present model system, that protons function
as mediators of generation of the R2 component of photo-
responses, can be successfully extended to the early

receptor potential of the visual photoreceptors.

217

The Protonic Induction of Ionic
Membrane ExcitatiOn

 

 

The mechanism of the generation of early receptor
potentials, presented in this work, is independent of the
organization of the membranous structure if there exist
the pigmented organelle which can generate protons and
the proton sensitive membrane electrode which has a
prOper proton permeability.

Since the early receptor potentials are most likely
to be membrane phenomena, it is possible to consider that
the transition from "non-ionic" phenomenon to "ionic"
phenomenon (electroretinogram) can take place across the
disc membrane, i.e. the electroretinograms may be
explained as the induced potentials of sodium, potassium,
and chloride ions by the initial rapid protonic potentials
(early receptor potentials) as shown in Figure 41.

In spite of enormous works on the determination of
the internal pH of excitable cells, there is some con-
troversy over not only the value of pH but also the effect
of pH gradient across the cell membrane on the induction
of excitation.(Caldwell, 1956;Wbddell and Bates, 1969).
Nevertheless, the intracellular pH is about 7 and the pH
gradient across the cell membrane is far less than 1 pH
unit (Caldwell, 1958, 1960). With an improved technique
Carter, Rector, Campion, and Seldin (1967) observed a
Donnan distribution of protons with instantaneous

response to an applied membrane potential. Furthermore,

218

Figure 41.--The schematic diagram of photoelectric
transduction showing how hydrogen ion
generation and transport can function as
the intermediate stages (early receptor
potentials) leading to the ultimate Na+
and K+ transport (the earlier component
of electroretinogram) across the photo-

receptor membrane.

219

 

-9 Pigment

 

 

 

 

\f

 

R1 of Early
Receptor Potential

 

Electron or Hole

 

 

Generation of the
Mediator of Early
Receptor Potential

R2 of Early
Receptor Potential

Earlier Components
of Electroretinogram

 

 

 

 

 

 

 

 

 

Transport
"-'—"'7
I
Proton I
Generation I
I
I
I
I
\/ I
I
Proton 1
Transport |
_.__._.J

 

\V

 

+
Na+ and K
Transport

 

 

 

Redox
Electrode

Hydrogen
Electrode

+
Na+ and K
Electrode

.’L‘L _ O
l

220

Tasaki, Singer, and Takenaka (1965) observed that pH of
the perfusion solution was essential for the production

of normal electrical discharges of squid giant axons:

the conduction was restored by changing the pH of the
bathing solution of the same salt composition but with
more alkaline pH, and at a pH greater than 7.5 the
spontaneous repetitive discharges were observed. Thus,

at the present it is not possible to eliminate the
importance of proton concentration in the induction of the
excitable membranes.

Recently, Stephens (1969) proposed that permeability
changes of nerve membranes could be initiated by the rapid
transport of protons, which control the chemical and
structural properties of channels for the transport of
cations. The influence of proton concentration upon the
ionic permeability of mitochondrial and chloroplast
membranes have been vigorously developed by Mitchell
(1961, 1966). Mitchell's chemiosomotic hypothesis is
not based on the structural changes of the membrane due
to changes in proton concentrations at the vicinity of
the membrane as is Stephens' hypothesis, but it is
essentially the coupling of the charge flow across the
membrane as previously suggested in terms of ionic
coupling by Lund (1928), Lundegardh (1945), and Ussing
(1947). It is clear from the above hypothesis that,

given the local disturbance on proton concentration at

221

the vicinity of the excitable membrane, the diffusion
of protons due to this concentration gradient can cause
the secondary flows of cations or anions across the
membrane to neutralize the charges by utilizing metabolic
energy (Mitchell's hypothesis) or by altering the
structure of the channels (Stephens' hypothesis).

The Effect of Temperature on Proton

ActiVity in the Induction
of’PhotopotenEiaIs

 

 

 

A successful demonstration of the biphasic photo-
response in an all-transfretinal BLM clearly indicates
that this model membrane can be used for the study of
the early receptor potential. It was further demon-
strated that the temperature dependence of the mode of the
generation of this biphasic photoresponse closely resembles
that of the early receptor potential. The similarities
between the early receptor potential and the photoresponse
of all-trans-retinal BLM as a function of temperature
can be seen in the following two points:

1. At the lower temperature the fast component

(R1) dominates the photoresponse while at the
higher temperature the slower component (R2)
dominates.

2. The shift from R1 to R2 can occur within the

temperature range from 0°C to 50°C under

proper experimental conditions.

222

It is appropriate to consider the agreement in the above
two points as the principal similarity between the two
kinds of the photoresponses.

Unlike the present membrane model, the effect of
temperature on the mode of the early receptor potential
can be the effect of temperature via two mechanisms. The
first mechanism is an increase in proton mobility as
demonstrated in the present work. The second one is the
direct effect of pH gradient across the receptor membrane,
since the effect of temperature on the pH of the aqueous
phases is a well established phenomenon in biological
samples. Lovelock (1957) points out that when dilute
aqueous phases are frozen, ice separates as a pure sub-
stance and the solutes are concentrated in the Spaces
remaining between the ice crystals. The increase in
electrolyte concentration may appear as a change in proton
concentration. Furthermore, sparingly soluble buffering
salts may come out of solutions as the temperature falls
and as water is removed from the solution.

The actual determination of pH changes due to
temperature changes have been carried out by Finn (1932)
and Harvey (1918) in muscle and cabbage juice respectively.
More recently, Van den Berg (1961a, b) observed changes
in the pH of milk, green beans, cauliflower, tomatoes,
cod, haddock, and beef. Most of these works were carried

out at subzero temperatures and the mode of the pH changes

223

is complex. Van den Berg felt that these changes in pH
were likely to be due to the precipitation of salts as
demonstrated with pure buffer systems (Van den Berg and
Rose, 1959a, b). Both such a precipitation of salts and
the conformation changes or denaturation of proteins due
to changes in pH would cause changes in the buffering
capacity of the intracellular and intercellular liquid
(Harvey, 1918). At the present time it is almost certain
that temperature influences the pH and buffer capacity of
the cellular liquid phases.

Finn (1932) determined changes in pH of muscle
juice at temperatures above freezing. Again, pH decreased
as temperatures were decreased. Since there are no
comparable data on photoreceptors or axons, a definite
statement cannot be made. It is likely, however, that
the type of pH changes which generate the proton concen-
tration gradient across the photoreceptor membrane can
influence the mode of generation of the early receptor
potential, as demonstrated clearly in the present work
in the case of the photoresponse of all-trans-retinal
BLM separating two aqueous phases which are different in
their ionic compositions.

Finally, the present work demonstrates that it is
possible in principle and feasible in View of available
evidence about the early receptor potentials, to explain
the effect of temperature on the mode of generation of

the early receptor potentials without considering the

224

photochemistry of rhodopsin and its photoproducts. Those
workers who subscribe to the rhodopsin derivative
hypothesis have reached the conclusion that the early
receptor potential is not in the main pathway of the
visual excitation (Pak, 1968). The present work,
however, provides a rational basis for constructing an
alternative mechanism of the generation of the early
receptor potential by taking account of the active
function of the membranous structure to its origin, and
the proton as the mediator of the generation of the early
receptor potential. Future works may demonstrate that
this approach will eventually elucidate the linkage
between the early receptor potential and the generation
3f the electroretinogram as coupled consecutive membrane

phenomena.

1. "'7.
1

CHAPTER VIII

SUMMARY

1. Ionic selectivity of all-trans-retinal BLM was
established: proton transport can satisfactorily account
for the develOpment of the observed dark membrane
potentials and no other ionic species or mechanisms are
required.

2. The time course of photopotentials showed a
biphasic mode with the temperature sensitivity which,
under apprOpriate experimental conditions, was quite
similar to that of early receptor pOtentials.

3. The initial fast photoresponse (R1) could be
best explained as a redox potential due to the photo—
excited electrons and holes which interact with water
molecules and/or other electron acceptors or donors.

This potential was not influenced by the concentration
gradient of K+, Na+, and C1- to any extent, indicating
the non-ionic origin of this photoresponse.

4. The second slow photoresponse (R2) was
identified as the diffusion potential of hydrogen ions.

5. The effect of temperature on the mode of
biphasic photoresponses was due to an increase in hydrogen

ion mobility with increasing temperature. It was discussed

225

226

that in vivo there could be another important factor of
the generation of the pH gradient across receptor cell
membranes as indicated by an increase in the R2 component
with increasing pH gradient, since there do exist reports
about the changes in pH of biological fluids as tempera-
ture was changed.

6. One of the future problems is to investigate
the detailed mechanism of generation of the R1 component,
as guided by the kinetics of the Langmuir adsorption
isotherm as developed in the present work.

7. From a biological viewpoint, an important prob-
lem may be whether or not this hydrogen ion movement
induced by the hole movement can induce, in turn, the
movement of Na+, K+, or Cl- ions across the membrane.
Since all-trans-retinal BLM was permeable only to
protons, this point could not be investigated. If
ionic (e.g., Na+, K+, or Cl_) selectivity can be
incorporated into this membrane system without lowering

the membrane resistance unduly, this three—step photo-

excitation potential may be observable.

r.

Mali)- (.1 .1...) [Alb-slit..- git!

BIBLIOGRAPHY

227

BI BLIOGRAPHY

Abrahamson, E. W., R. Adams, and V. I. Wulff (1959),
"Reversible spectral changes in retinene solutions
following flash illumination," J. Phys. Chem. 63,
441-443. '

Abrahamson, E. W. and S. E. Ostroy (1967), "The photo-
chemical and macromolecular aspects of vision,"
Progr. Bithys. 17, 179-215.

Adams, R. G. and W. A. Hagins (1960), "The ionic compo-
sition of squid photoreceptors," Biol. Bull. 119,
300-301.

 

Anderson, 0. R., O. A. Roels, K. D. Dreher, and J. H.
Schulman (1967), "The stability and structure of
mixed lipid monolayers and bilayers II. The effect
of retinol and a-tocopherol on the structure and
stability of lipid bilayers," J. Ultrastruct. Res.
19, 600—610. _

 

Arden, G. B., C. D. B. Bridges, H. Ikeda, and I. M. Siegel
(1966a), "Isolation of a new fast component of the
early receptor potential," 1. Physiol. 186, 123p-124p.

Arden, G. B., C. D. B. Bridges, H. Ikeda, and I. M. Siegel
(1966b), "Rapid light-induced potentials common to
plant and animal tissues," Nature 212, 1235-1236.

Arden, G. B., C. D. B. Bridges, H. Ikeda, and I. M. Siegel
(1968), "Mode of generation of the early receptor
potential,” Vision Res. 8, 3-24.

 

Arden. G. B. and H. Ikeda (1965), "A new property of the
early receptor potential of rat retina," Nature
208, 1100-1101.

Arden, G. B. and H. Ikeda (1966), "Effects of hereditary
degeneration of the retina on the early receptor
potential and the corneo-fundal potential of the
rat eye," Vision Res. 6, 171-184.

 

Arden, G. B., H. Ikeda, and M. Siegel (1966a), "New
components of the mammalian receptor potential and
their relation to visual photochemistry," Vision
Res. 6, 373-384.

228

229

Arden, G. B., H. Ikeda, and M. Siegel (1966b), "Effects of
light—adaptation on the early receptor potential,"
Vision Res. 6, 357-371.

 

Azuma, K. and M. Takagi (1966), "Rhodopsin film at air-
water interface," Ann. Rept. Biol. Works Fac. Sci.
Osaka Univ. 14, 73-81.

 

 

 

 

Bangham, A. D. (1968), "Membrane models with phospholipids,"
Progr. BiOphys. Molec. Biol. 18, 29—95.

Bangham, A. D., J. T. Dingle, and J. A. Lucy (1964), q 1
"Studies on the mode of action of excess of vitamin A H
A: 9. Penetration of lipid monolayers by compounds
in the vitamin A series," Biochem. J. 90, 133-140.

Becker, H. E. and R. A. Cone (1966), "Light-stimulated

electrical responses from skin," Science 154, 1051- 9 _
1053. A _w

Beyerle, M. E. and H. Gerischer (1967), "Uber die Roller
der Protonen bei der Injektion von Defektelektronen
in Anthracenkristallen," Ber. Bunsenges. Phisik.
Chem. 71, 380-383.

 

 

Blasie, J. K. and C. R. Worthington (1969), "Planar
liquid-like arrangement of photOpigment molecules
in frog retinal receptor disk membranes," J. M31.
Biol. 39, 417-439. ‘

Blaurock, A. E. and M. H. F. Wilkins (1969), "Structure

of frog photoreceptor membranes," Nature 223, 906-
909.

Bonting, S. L. and A. D. Bangham (1967), "On the bio-
chemical mechanism of the visual process," Exptl,
Eye Res. 6, 400-413.

Brindley, G. S. and A. R. Gardner—Medwin (1965), "Evidence
that the 'early receptor potential' of the retina
does not depend on changes in the permeability of
membranes to specific ions," J. Physiol. 180, 1p.

Brindley, G. S. and A. R. Gardner-Medwin (1966), "The
origin of the early receptor potential of the
retina," J. Physiol. 182, 185—194.

Broda, E. E. and E. Victor (1940), "The cataphoretic
mobility of visual purple," Biochem. J. 34, 1501-
1506.

230

Brown, K. T. (1965), "An early potential evoked by light
from the pigment epithelium—choroid complex of the
eye of the toad," Nature 207, 1249-1253.

Brown, K. T. and J. M. Crawford (1966), "Intracellular
recording of rapid light-evoked responses from
pigment epithelium cells of the frog eye,"
Physiologists 9, 146.

 

Brown, K. T. and J. M. Crawford (1967a), "Intracellular
recording of rapid light—evoked responses from
pigment epithelium cells of the frog eye," Vision
Res. 7, 149-163.

Brown, K. T. and J. M. Crawford (1967b), "Melanin and the
rapid light-evoked responses from pigment epithelium
cells of the frog eye," Vision Res. 7, 165-178.

 

Brown, K. T. and P. W. Gage (1966), "An earlier phase of
the light—evoked electrical response from the
pigment epithelium-choroid complex of the eye of
the toad," Nature 211, 155—158.

Brown, K.’T.enxibL Murakami (1964aL"A.new receptor potential
of the monkey retina with no detectable latency,"
Nature 201, 626—628.

Brown, K. T. and M. Murakami (1964b), "Biphasic form of
the early receptor potential of the monkey retina,"
Nature 204, 739-740.

Brown, K. T., K. Watanabe, and M. Murakami (1965), "The
early and late receptor potentials of monkey cones
and rods," Cold Spring Harbor Symp. Quant. Biol.
30, 457-482.

 

 

Brown, K. T. and T. N. Wiesel (1961), "Localization of
origins of electroretinogram components by intra-
retinal recording in the intact cat eye," J.
Physiol. 158, 257-280. —

Caldwell, P. C. (1956), "Intracellular pH," Int'l. Rev.
Cytol. 5, 229-277.

 

Caldwell, P. C. (1958), "Studies on the internal pH of
large muscle and nerve fibers," J. Physiol. 142,
22—62. _

Carter, N. W., F. C. Rector, D. S. Campion, and D. W.
Seldin (1967), "Measurement of intracellular pH of
skeletal muscle with pH-sensitive glass micro—
electrodes," 1. Clin. Invest. 46, 920-933.

 

231

Castleden, J. A. (1969), "Phospholipid cell-membrane
models," J. Pharm. Sci. 58, 149-165.

 

Chynoweth, A. G. and W. G. Schneider (1954), "The
photoconductivity of anthracene I," J. Chem. Phy .
22, 1021-1028.

 

Cohen, A. I. (1968), "New evidence supporting the
linkage to extracellular space of outer segment
saccules of frog cones but not rods," 1. Cell Biol.
37, 424-444.

 

Colmano, G. (1962), "Visible absorption spectrum of
chlorophyll a, b, and B-carotene molecules mixed
in monolayers at a water-air interface," Nature
193, 1287-1288.

Colmano, G and S. McGlone (1961), "B-carotene films at a
water-air interface," Biochim. Biophys. Acta 53,
429-431.

 

Cone, R. A. (1964), "Early receptor potential of the
vertebrate retina," Nature 204, 736-739.

Cone, R. A. (1965), "The early receptor potential of the
vertebrate eye," Cold Spring Harbor Symp. Quant.
Biol. 30, 483-491.

 

 

Cone, R. A. (1967), "Early receptor potentials: Photo-
reversible charge displacement in rhodopsin,"
Science 155, 1128-1131.

Cone, R. A. and P. K. Brown (1967), "Dependence of the
early receptor potential on the orientation of
rhodOpsin," Science 156, 536.

Cone, R. A. and W. H. Cobbs (1969), "RhodOpsin cycle in
the living eye of the rat," Nature 221, 820-822.

Conway, E. J. (1953), "A redox pump for the biological
performance of osmotic work, and its relation to
the kinetics of free ion diffusion across membranes,’
Int'l. Rev. Cytol. 2, 419-445.

 

 

Crane, E. E., R. E. Davies, and N. M. Longmuir (1948a),
"Relations between hydrochloric acid secretion and
electrical phenomena in frog gastric mucosa,"
Biochem, J. 43, 321-336.

232

Crane, E. E., R. E. Davies, and N. M. Longmuir (1948b),
"The effect of electric current on HCl secretion

by isolated frog gastric mucosa," Biochem. J. 43,
336-342. ""“" ‘ '

Crawford, J. M., P. W. Gage, and K. T. Brown (1967),
"Rapid light-evoked potentials at extremes of pH
from the frog's retina and pigment epithelium,
and from a synthetic melanin," Vision Res. 7, 539-
551.

 

Danielli, J. F. and H. Davson (1935), "A contribution to
the theory of permeability of thin films," J. Cell
Comp. Physiol. 5, 495-508.

 

 

Dartnall, H. J. A. (1948), "Indicator yellow and
retinenel," Nature 162, 222.

Davies, R. E. and A. G. Obston (1950), "On the mechanism
of secretion of ions by gastric mucosa and by other
tissues," Biochem. J. 46, 324-333.

Dawson, W. and E. W. Abrahamson (1962), "Population and
decay of the lowest triplet state in polyenes with
conjugated heteroatoms: retinene," J. Phys. Chem.
66, 2542-2547. ‘

 

Dewey, M. M., P. K. Davis, J. K. Blasie, and L. Barr
(1969), "Localization of rhodopsin antibody in
retina of the frog," J. Mol. Biol. 39, 395-405.

 

Dowling, J. E. (1967), "The organization of vertebrate
visual receptors," in Molecular Organization and
Biological Function, J. M. Allen, ed., Harper and
Row, New York, pp. 186-210.

 

 

Dreher, K. D., J. H. Schulman, O. R. Anderson, and O. A.
Roels (1967), "The stability and structure of mixed
lipid monolayers and bilayers I. Properties of
lipidand lipoproteinrmnmflayers on 0504 solutions
and the role of cholesterol, retinol, and tocopherol
in stabilizing lecithin monolayers," J. Ultrastruct.
Res. 19, 586-599.

 

Ebrey, T. G. (1967), "Fast light-evoked potential from
leaves," Science 155, 1556-1557.

Ebrey, T. G. (1968), "Characterization of the fast light-
evoked electrical responses from leaves," Photochem.
Photobiol. 7, 297-309.

 

 

233

Eguchi, E. and T. H. Waterman (1966), "Fine structure
patterns in crustacean rhabdoms," in The Functional
Organization of the Compound Eye, C. G. Bernhard,
ed., Pergamon—Press, New York, pp. 105-124.

 

 

 

Eigen, M. and L. De Maeyer (1958), "Self-dissociation and
protonic charge transport in water and ice," Proc.
Royal Soc. London A247, 505-533.

 

Falk, G. and P. Fatt (1966), "Rapid hydrogen ion uptake
of rod outer segments and rhodopsin solutions on
illumination," J. Physiol. 183, 211-224.

Feldman, G. L. (1967), "The lipids in pathology of the
eye," J. Am. Oil Chemists' Soc. 44, 615-622.

 

Finean, J. B. (1969), "Biophysical contributions to
membrane structure," Quart. Rev. Biophys. 2, 1-23.

Finean, J. B. and P. F. Millington (1955), "Low-angle
X-ray diffraction study of the polymorphic forms
of synthetic a:B and d:d' kephalins and d:B
lecithins," Trans. Faradanyoc. 51, 1008-1015.

 

 

Finn, D. B. (1932), "Denaturation of proteins in muscle
juice by freezing," Proc. Royal Soc. London B111,
396-411.

 

Fitzpatrick, T. B., W. Quevedo, A. L. Levene, F. J.
McGovern, Y. Mishima, and A. G. Oettle (1966),
"Terminology of vertebrate melanin-containing cells:
1965," Science 152, 88-89.

Fujishita, S. (1962), "The organo-iron compound in the
digitonin extract of the rod outer segment,"
Jap. J. Physiol. 152, 460-466.

Fujishita, S. (1968), "Changes in pH of rhodopsin solution
on exposure to a dim green light," Jap. J. Physiol.
18, 303-309.

Fujishita, S. (1969), "Hydrogen ion uptake by oxygen
atoms released from rhodopsin molecules on
illumination," Jap. J. Physiol. 19, 455-464.

Fukami, I., B. L. Vallee, and G. Wald (1959), "Does
rhodopsin contain a trace metal?" Nature 183,
28-30.

Gage, P. W., J. M. Crawford, and K. T. Brown (1968),
"Ionic dependence of the third phase of the light-
evoked response from amphibian pigment epithelium
cells," Vision Res. 8, 369-382.

 

234

Galloway, N. R. (1967), "Early receptor potential in the
human eye," Brit. J. Opthal. 51, 261-264.

Goldstein, E. B. (1967), "Early receptor potential of the
isolated frog (Rana Pipiens) retina," Vision Res.
7, 837-845.

 

Goldstein, E. B. (1968), "Visual pigments and the early
receptor potential of the isolated frog retina,"
Vision Res. 8, 953-963.

 

Goldstein, E. B. (1969), "Contribution of cones to the
early receptor potential in the rhesus monkey,"
Nature 222, 1273-1274.

Goldstein, E. B. and E. L. Berson (1969), "Cone dominance
of the human early receptor potential," Nature 222,
1272—1273.

Goldup, A., S. Ohki, and J. F. Danielli (1970), "Black
lipid films," Progr. Surface Sci. 3, 193-260.

 

Gorter, E. and F. Grendel (1925), "On bimolecular layers
of lipids on the chromocytes of the blood," J.
Exptl. Med. 41, 439-443.

 

Grellmann, K., R. Menning, and R. Livingston (1962),
"Some flash-photolytic and photochemical studies
of retinene and related compounds," J. Am. Chem.
Soc. 84, 546-548. — “—

 

Gurney, R. W. (1962), Ions in_Solution, Dover Publication,
Inc., New York, p. 3.

Hagins, W. A. and R. E. McGaughy (1967), "Molecular and
thermal origins of fast photoelectric effects in
the squid retina," Science 157, 813-816.

Hagins, W. A. and R. E. McGaughy (1968), "Membrane
origin of the fast photovoltage of squid retina,"
Science 159, 213-215.

Hanai, T., D. A. Haydon, and J. Taylor (1964), "An
investigation by electrical methods of lecithin-in-
hydrocarbon films in aqueous solutions," Proc.
Royal Soc. London A281, 377-391.

 

Hanai, T., D. A. Haydon, and J. Taylor (1965), "Polar
group orientation and the electrical prOperties
of lecithin bimolecular leaflets," J. Theoret. Biol.
9, 278-296. ' “

235

Hara, R. and T. Hara (1960), "Changes in electrical con-
ductance of rhodopsin on exposure to flashlight,"
Zool. Mag., Tokyo 69, 262-268.

 

Hara, T. (1958), "The effect of illumination on the electri-
cal conduction of rhodopsin solutions," J. Gen.
Physiol. 41, 857-877.

Hartley, G. S. and J. W. Roe (1940), "Ionic concentrations
at interfaces," Trans. Faraday Soc. 36, 101-109.

 

Harvey, R. B. (1918), "Hardening process in plants and devel-
Opments from frost injury," J. Agr. Res. 15, 83-111.

Hempel, K. (1966), "Investigation on the structure of
melanin in malignant melanoma with 3H- and l4C-d0pa
labeled at different positions," in Structure and
Control of the Melanogytes, G. Della Porta and O.
MUhlBock7—eds., Sprifiger, New York, pp. 162-175.

 

 

Henn, F. A. and T. E. Thompson (1968), "PrOperties of lipid
bilayer membranes separating two aqueous phases:
composition studies," J. Mol. Biol. 31, 227-235.

Henn, F. A. and T. E. Thompson (1969), "Synthetic lipid
bilayer membranes," Ann. Rev. Biochem. 38, 241-262.

Hesketh, T. R. (1969), "Photoconductivity in black lipid
and thin lipid membranes," Nature 224, 1026-1028.

Hyono, A., S. Kuriyama, K. Tsuji, and Y. Hosoya (1962),
"Monolayer film of rhodopsin at the air/water inter-
face," Nature 193, 679-680.

Jahn, T. L. (1962), "A theory of electronic conduction
through membranes, and of active transport of ions,
based on redox transmembrane potentials," J. Theoret.
Biol. 2, 129-138. '

Jahn, T. L. (1963), "A possible mechanism for the amplifier
effect in the retina," Vision Res. 3, 25-28.

 

Kallmann, H. and M. Pope (1960), "Positive hole injection
into organic crystals," J. Chem. Phys. 32, 300-301.

 

Karyakin, A. V., G. A. Muradova, and G. N. Sayenko (1969),
"State of water in substances taking part in the
process of photosynthesis," Biofizika 14, 240-244.

 

Kasha, M. (1948), "Transmission filters for the ultra-
violet," J. Opt. Soc. Am. 38, 929-934.

236

Kay, R. E. and H. Chan (1969), "Photoelectric responses
of lipid membranes," Radiat. Res. 40, 177-192.

 

Kobakova, E. M. (1946), J. Physiol. U.S.S.R. 32, 385;
quoted by A. Pirie_and R. Van Heyninger,
Biochemistry of the Eye, Charles C. Thomas,
Springfield: IIlTT_l956, pp. 213-214.

 

 

Langer, H. and B. Thorell (1966), "Microspectrophotometric
assay of visual pigments in single rhabdomeres of
the insect eye," in The Functional Organization of
the Compound Eye, C. G. Bernhard, ed}, Pergamon
Press, New York, pp. 145-149.

 

 

Langmuir, I. (1918), "The adsorption of gases on plane
surfaces of glass, mica and platinum," J. Amer.
Chem. Soc. 40, 1361-1403.

 

Langmuir, I. (1921), "Chemical reactions on surfaces,"
Trans. Faraday Soc. 17, 607-620.

 

 

Lasansky, A. and M. G. F. Fuortes (1969), "The site of
origin of electrical responses in visual cells
of the leech, Hirudo Medicinalis," J. Cell Biol.
42, 241-252. _

 

 

 

Leslie, R. B. and D. Chapman (1967), "Artificial phospho-
lipid membranes and bioenergetics," Chem. Phys.
Lipids 1, 143-156.

 

Lettvin, J. Y. (1965), "General discussion: early
receptor potential," Cold Spring Harbor Symp.
Quant. Biol. 30, 501-502.

 

 

Levi, H. and H. H. Ussing (1948), "The exchange of
sodium and chloride ions across the fibre membrane
of the isolated frog sartorium," Acta Physiol.
Scand. 16, 232-249.

 

 

Lillie, R. S. (1915), "The conditions of conduction of
excitation in irritable cells and tissues and
especially in nerve 1," Am. J. Physiol. 37,
348-370.

Lillie, R. S. (1916), "The conditions of physiological
conduction in irritable tissues III. Electrolytic
local action as the basis of propagation of the
excitation-wave,” Am. J. Physiol. 41, 126-136.

Lovelock, J. E. (1957), "The denaturation of lipoid-
protein complexes as a cause of damage by freezing,"
Proc. Royal Soc. London B147, 427-433.

 

 

237

Lund, E. J. (1928), "Relation between continuous bio-
electric current and cell respiration V. The quanti-
tative relation between E and cell oxidation as
shown by the effect of Cyanide and oxygen," J. Exp.
Zool. 51, 327-337.

Lundegardh, H. (1939), "An electro-chemical theory of salt
adsorption and respiration," Nature 143, 203-204.

Lundegardh, H. (1945), "Absorption, transport and exudation
of inorganic ions by the roots," Arkiv. Bot. A32, 1-139.

Lundegardh, H. (1946), "Transport of water and salts
through plant tissues," Nature 157, 575-577.

MacInnes, D. A. (1961), The Principles pf Electrochemistry,
Dover Publications, Inc., New York, p. 225.

 

 

Maeda, Y. and T. Isemura (1967), "Monolayer films of retinal
and the effects of light on them," Nature 215, 765-766.

Marsh, G. (1935), "Kinetics of an intracellular system for
respiration and bioelectric potential at flux equi-
librium," Plant Physiol. 10, 681-697.

 

Mason, H. S. (1967), "The structure of melanin," Adv. Biol.
Skin 8, 293-312.

 

Mauzerall, D. and A. Finkelstein (1969), "Light induced
changes in the conductivity of thin lipid membranes
in the presence of iodine and iodide ion," Nature 224,
690-692.

McConnell, D. G. (1965), "The isolation of retinal outer
segment fragments," J. Cell Biol. 27, 459-473.

 

McConnell, D. G., C. N. Rafferty, and R. A. Dilley (1968),
"The light-induced proton uptake in bovine retinal
outer segment fragments," J. Biol. Chem. 25, 5820-5826.

Mitchell, P. (1961), "Coupling of phosphorylation to electron
and hydrogen transfer by chemiosomotic type of
mechanism," Nature 191, 144-148.

Mitchell, P. (1966), "Chemiosomotic coupling in oxidative and
photosynthetic phosphorylation," Biol. Rev. 41, 445-502.

 

Miyamoto, V. K. (1966), 2. 9. Electrical Prpperties pf Lipid
Bilayer Membranes, Ph.D. Thesis, The Johns Hopkins
University, Baltimore, Md.

 

238

Mueller, P. and D. O. Rudin (1969), "Translators in bimolecu-
lar lipid membranes: their role in dissipative and
conservative bioenergy transductions," Curr. TOpics
Bioenergetics 3, 157-249.

 

 

Mueller, P., D. O. Rudin, H. T. Tien, and W. C. Wescott
(1962a), "Reconstitution of cell membrane structure
in vitro and its transformation into an excitable

system,‘ Nature 194, 979-980.

Mueller, P., D. O. Rudin, H. T. Tien, and W. C. Wescott
(1962b), "Reconstruction of excitable cell membrane
structure ip_vitro," Circulation 26, 1167-1171.

 

Mueller, P., D. O. Rudin, H. T. Tien, and W. C. Wescott
(1963), "Methods for the formation of single
bimolecular lipid membranes in aqueous solution,"
J. Phys. Chem. 67, 534-535.

 

Mueller, P., D. O. Rudin, H. T. Tien, and W. C. Wescott
(1964), "Formation and properties of bimolecular
lipid membranes," Recent Progr. Surface Sci. 1,
379-393.

 

Ostrovskii, M. A., I. B. Fedorovich, and S. Y. Polyak
(1968), "Change in pH of medium on illumination of
the retina and a suspension of the external segment
of photoreceptors," Biofizika 13, 338-339.

 

Pak, W. L. (1965), "Some properties of the early electrical
response in the vertebrate retina," Cold Spring
Harbor Symp. Quant. Biol. 30, 493-499.

 

Pak, W. L. (1968), "Rapid photoresponses in the retina
and their relevance to vision research," Photochem.
Photobiol. 8, 495-503.

 

 

Pak, W. L. and R. J. Boez (1967), "Rhodopsin: responses
from transient intermediates formed during its
bleaching," Science 155, 1131-1133.

Pak, W. L. and R. A. Cone (1964), "Isolation and identifi-
cation of the initial peak of the early receptor
potential," Nature 204-836-838.

Pak, W. L. and T. G. Ebrey (1965), "Visual receptor
potential observed at sub-zero temperatures,"
Nature 205, 484-486.

Pak, W. L. and T. G. Ebrey (1966), "Early receptor
potentials of rods and cones in rodents," J. Gen.
Physiol. 49, 1199-1208.

239

Pak, W. L., V. P. Rozzi, and T. G. Ebrey (1967), "Effect
of changes in the chemical environment of the
retina on the two components of the early receptor
potential," Nature 214, 109-110.

Pant, H. C. and B. Rosenberg (1970), "Photochemistry and
photo-electric effects in bimolecular lipid membranes
in presence of inorganic light sensitive ions,"
Biophys. S93. Abst. 10, 70a.

Pedler, C. M. H. and R. Tilly (1967), "The fine structure
of photoreceptor discs," Vision Res. 7, 829-836.

 

Radding, C. M. and G. Wald (l955-l956a), "Acid-base
properties of rhodopsin and opsin," J. Gen. Physiol.
39, 909-922.

Radding, C. M. and G. Wald (l955-l956b), "The stability
of rhodopsin and Opsin. Effects of pH and aging,"
1. Gen. PhYSiOlo 39' 923-9330

 

Reeves, J. P. (1970), "Model membrane systems,‘ in
Biological Membranes, R. M. Dowben, ed., Little,
Brown, Boston, pp. 223-254.

 

Ried, C. (1957), Excited States £3 Chemistry and Biology,
Academic Press, Inc., New York, p. 162.

 

Robertson, R. N. and M. J. Wilkins (1948a), "Quantitative
relation between salt accumulation and salt
respiration in plant cells," Nature 161, 101.

Robertson, R. N. and M. J. Wilkins (1948b), "Studies in
the metabolism of plant cells. VII. The quantita-
tive relation between salt accumulation and salt
respiration," Aust. J. Sgi. Egg. Bl, 17-37.

Roels, O. A. and D. O. Shah (1969), "Molecular inter-
actions in lecithin-retinol monolayers," J.
Colloid Interface Sci. 29, 279-287.

 

Rosenberg, B. (1958), "Photoconductivity and the visual
receptor process," J. Opt. Soc. Am. 48, 581-582.

Rosenberg, B. and H. C. Harder (1967), "Semiconduction
and photoconduction activation energy of the
retinals," Photochem. Photobiol. 6, 629-641.

240

Rosenberg, B., R. A. Orlando, and J. M. Orlando (1961),
"Photoconduction and semiconduction in dried

receptors of sheep eyes," Arc. Biochem. Biophys.
93, 395-398.

 

 

Rosenberg, B. and H. C. Pant (1969), "Photo-chemical and
photo-electric effects in an oxidized cholesterol
bimolecular lipid membrane in the presence of ferric
ions," Third Int'l. Biophys. Congr. Int'l. Union
Pure Appl. Biophys. Abst., 67. ——-_-' ——__—

 

 

Rosenberg, B. and E. Postow (1970), "Semiconduction in
protein and lipids--its possible biological import,"
Ann. New York Acad. Sci. 158, 161-190.

 

Rothfield, L. and A. Finkelstein (1968), "Membrane
biochemistry," Ann Rev. Biochem. 37, 463-496.

 

Schartzberg, P. (1965), "Diffusion of water through
hydrocarbon liquids," J. Polymer Sci, C10, 88-92.

Seiji, M., K. Shimao, M. S. C. Birbeck, and T. B.
Fitzpatrick (1963), "Subcellular localization of
melanin biosynthesis," Ann. New York Acad. Sci.
100, 497-533. “‘ —"

 

Smith, T. G. and J. E. Brown (1966), "A photoelectric
potential in invertebrate cells," Nature 212,
1217-1219.

Stephens, W. G. S. (1969), "Hydrogen ion and the
activation of electrically excitable membranes,"
Nature 224, 547-549.

Stiehler, R. D. and L. B. Flexner (1938), "A mechanism of
secretion in the chorioid plexus. The conversion

of oxidation-reduction energy into work," J. Biol.
Chem. 126, 603-617.

Takagi, M., K. Azuma, and U. Kishimoto (1965), "A new
method of bilayer membranes in aqueous solution,"
Ann. Rept. Biol. Works Fac. Sci. Osaka Univ. 13,
167-110. “'—

 

 

 

 

Tasaki, I., I. Singer, and T. Takenaka (1965), "Effects
of internal and external ionic environment on
excitability of squid giant axon: a macromolecular
approach," J. Egg. Physiol. 48, 1095-1123.

Taylor, H. S. (1938), "The temperature variation of
diffusion processes," J. Chem. Phys. 6, 331-334.

 

 

241

Tien, H. T. (1967), "Black lipid membranes: thickness
determination and molecular organization by
optical methods," J. Theoret. Biol. 16, 97-110.

 

Tien, H. T. (1968a), "Light-induced phenomena in black
lipid membranes constituted from photosynthetic
pigments," Nature 219, 272-274.

Tien, H. T. (1968b), "Photoelectric effects in thin and
{ bilayer lipid membranes in aqueous media," J. Phys.
Chem. 72, 4512-4519.

Tien, H. T. (1968c), "Black lipid membranes at bifaces:
formation characteristics, Optical and some
thermodynamic properties," J. Gen. Physiol. 52,
1255-1445. "

 

Tien, H. T. (1968d), Unpublished Results.

Tien, H. T., S. Carbone, and E. A. Dawidowicz (1966),
"Formation of 'black'lipid membranes by oxidation
products of cholesterol," Nature 212, 718-719.

Tien, H. T. and A. L. Diana (1967), "Black lipid
membranes in aqueous media: the effect of salts on

electrical properties," J. Colloid Interface Sci.
24, 287-296.

 

Tien, H. T. and A. L. Diana (1968), "Bimolecular lipid
membranes: a review and a summary of some recent
studies," Chem. Phys. Lipids 2, 55-101.

 

Tien, H. T. and N. Kobamoto (1969), "Carotenoid bilayer
lipid membrane model for the visual receptor,"
Nature 224, 1107-1108.

Tien, H. T. and S. P. Verma (1970), "Electronic processes
in bilayer lipid membranes," Nature 227, 1232-1234.

Ussing, H. H. (1947), "Interpretation of the exchange of
radio-sodium in isolated muscle," Nature 160,
262-263.

Van den Berg, L. (1961a), "Changes in pH of milk during
freezing and frozen storage," J. Dairy Res. 44,
26—31.

 

Van den Berg, L. (1961b), "Changes in pH of some frozen
foods during storage," Food Tech. 15, 434-437.

 

242

Van den Berg, L. and D. Rose (1959a), "Effect of freezing
on the pH and composition of sodium and potassium

 

phosphate solutions: the reciprocal system
KHZPO4-Na2HPO4-H20," Arch. Biochem. Biophys. 81,
319-329.

Van den Berg, L. and D. Rose (1959b), "The effect of
addition of sodium and potassium chloride to the
reciprocal system: KHzPO4-Na2HPO4-H20 on pH and
composition during freezing," Arch. Biochem. Biophys.
84, 305-315.

 

 

Waddell, W. J. and R. G. Bates (1969), "Intracellular pH,"
Physiol. Rev. 49, 285-329.

 

Wald, G. (1968), "The molecular basis of visual excitation,"
Nature 219, 800-807.

Walther, J. B. (1966), "Single cell responses from the
pigmentive eyes of an annelid," in Functional
Organization of the Compound E e, C. G. Bernhard,
ed., Pergamon—Press, New York, 29-336.

 

 

 

Wolken, J. J. (1961), "A structural model for a retinal
rod," in The Structure pf the E e, G. K. Smelser,
ed., Academic Press, New York, pp. 173-191.

 

Wolken, J. J. (1962), "Photoreceptor structure: their
molecular organization for energy transfer,"
J. Theoret. Biol. 3, 192-208.

 

Wolken, J. J. (1966), Vision: Biophysics and Biochemistry
pf the Retinal Photoreceptors, Thomas, Springfield,
111., p. 142.

 

 

Yonemura, D. and K. Kawasaki (1967), "The early receptor
potential in the human electroretinogram," Jap:
J. Physiol. 17, 235-244.

.nn—u-n ‘,l‘_'-: ’I‘V‘. -.l 1%..” . l-l'.‘
I.

        

 
 

YLBR

“((1)1111((1)1111 u 1'

9279

ARIES