|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 59
Phototransduction in Vertebrate Rods:
The Electrophysiological Approach to
the cGMP Cascade Theory
EDWARD N. PUGH, JR., W.H. CORES, AND ].W. TANNER
This paper reviews recent developments in the field of phototransduc-
tion, with a special focus on electrophysiological work that constrains and
informs the transduction theory now called the cyclic GMP cascade theory
(for other recent reviews, see Lamb, 1986; Pugh and Cobbs, 1986; Stryer,
1986; Yau et al., 1986; Pugh and Miller, 1987~.
INTrRODUCTION
The Nature of Phototransduction
Phototransduction may be defined as the sequence of events that tran-
spire in a photoreceptor from the absorption of a photon to the production
of a physiologically significant electrical signal. The latter ideally would be
specified as a light-induced membrane polarization at the synaptic region
sufficient to alter transmitter release to a degree detectable by a subjacent
bipolar cell, but practically can be equally well specified as a light-induced
change in cell membrane potential or even in outer-segment membrane
current.
Research in phototransduction took an exciting turn in 1970 with
the publication of Hagins et al.'s (1970) landmark paper on the dark
currents and photocurrents of rat rods. Figure 1 summarizes their finding,
overlaying the theoretical interpretation of their data on a scanning electron
micrograph of a frog retina. In the dark there is a continually flowing,
net outer-segment-inward membrane current, whose spatially integrated
magnitude is about 50 to 70 pA in the 25-pm outer-segment rat rod at
37°C (rat rods are essentially identical to human rods) and is about 40 to
50 pA in the 50-pm frog or toad rod. A little appreciation of the magnitude
59
OCR for page 60
EN PUGH' JR., W.H. CORES, ID ID. TONER
FIGURE 1 A scanning electron micrograph of the photoreceptor layer of the frog retina
on which has been overlaid a schematic of the "dark" or "circulating" current of a rod and
the ionic species that carry the current. SOURCE: Courtesy of W.H. Miller.
Of the dark current can be had from considering some numbers: since 1 pA
of current corresponds to 6.25 x 106 monovalent charges/second, a 50-pA
dark current flowing into a human rod outer segment with a water volume
of c. 100 fl demands a complete turnover of the ions carrying the current
every 5 seconds (50 pC of monovalent charge into 100 fl corresponds to
5 mM of the charge carrier; we can assume that the internal c[Na] is
no higher than 50 mM). For the rod membrane potential to be at steady
state in the face of the inward current, there must be a balancing outward
current in the inner segment. General principles of cellular physiology,
ion substitution experiments, and other work lead to the conclusions that
60
OCR for page 61
PHOTOTRANSDUCTION IN VERTEBRATE RODS
61
(1) the outer-segment inward current is carried primarily by Na+, (2) the
balancing inner-segment outward current is carried primarily by K+, and
(3) there exists a potent ATP-requiring Na/K exchange that maintains the
dark ionic balance. The latter exchange mechanism has been localized by
Stirling and Lee (1980) to be primarily in the ellipsoid region and on the
calices of inner-segment membrane that climb up the inner part of the
outer segment.
Hagins et al. (1970) showed that transverse flashes of light delivered
to the outer-segment layer produce local membrane currents photocur-
rents—and hypothesized that the photocurrent was merely the suppression
of the dark current. Penn and Hagins (1972) showed that the maximal
photocurrent was in fact the complete suppression of the dark current. That
the rod should hyperpolarize to light follows immediately from the fact that
light suppresses the outer-segment dark current: since that latter current
is a depolarizing current, its suppression (in the absence of compensatory
reaction) must necessarily drive the membrane potential toward the reversal
potential of any K+ conductance. Thus, by 1970 the nature of the initial
electrophysiological event in phototransduction was made crystal clear: it
is the local suppression in the outer segment of inward membrane current.
Two Fundamental Constraints
We would now like to emphasize two additional hard inferences drawn
in the early 1970s, inferences that drove research on the theory of photo-
transduction through much of the 1970s and early 1980s. The first was that
there must exist a specific conductance in the outer-segment membrane
that carries the dark current and that is closed by light the light-sensitive
conductance, or go. Hagins et al.'s (1970) work and subsequent work with
the suction electrode (Baylor et al., 1979; Lamb et al., 1981) showed ghu
to be distributed more or less uniformly throughout the outer-segment
membrane and also showed that the effect of a single photon on the
conductance was highly localized, perhaps to 2 to 3 ~m. A second hard
inference that propelled transduction research was that there must exist an
internal messenger in rods.
Light-Sensitive Conductance
The inference that there must be a specific light-sensitive conductance
motivated many "fingerprint" analyses: experiments for characterizing this
entity operationally. Chief among these are currentholtage relations, noise
power spectra, and ionic permeation experiments. Biophysical character-
~zation might be said to have begun with Bader et al.'s (1979) effort to
OCR for page 62
62
E.N PUGH, JR, W.~. CORES, AND ID. TONER
determine the currentivoltage (I/V) relation of the conductance; the re-
lation was definitively determined in the elegant triple electrode study of
Baylor and Nunn (1986~. One striking feature of this I/V relation is its
near-zero slope over the entire physiological range of voltages, - 25 mV to
—75 mV. As Baylor has pointed out, this unusual I/V relation is ideal for
producing virtually no cable losses in the outer segment, thus allowing the
full effect of a local dark-current suppression to be converted into a voltage
change in the inner-segment membrane. Another basic characterization
of gin, was carried out by Bodoia and Detwiler (1984), who measured
noise power spectra that could be attributed to noise produced by the
conductance per se. Baylor et al. (1980) had previously characterized two
other forms of outer-segment noise that manifest themselves through the
light-sensitive conductance: (1) "shot noise" events that can, with virtual
certainty, be attributed to the thermal isomerization of rhodopsin and (2) a
broader band noise that corresponds to some as yet unspecified transducer
process.
Internal Messenger Hypothesis
The second inference that drove much transduction work was the need
for an internal messenger, required because of several different kinds of
evidence that the rod disks (which one could presume are the site of most
photon adsorptions, based on the action spectrum of night vision) are
physically separated from the outer-segment plasma membrane by about
the distance of a synaptic cleft. In 1970 Yoshikami and Hagins, in analogy
with other systems such as muscle, proposed the divalent cation Ca2+ to be
the transmitter. About the same time Bitensky et al. (1971), in analogy with
Sutherland's fundamental insight, proposed that a cyclic nucleotide was the
messenger. It is not our goal to review the incredible burgeoning field of
cyclic nucleotide research that followed the Miller and Bitensky hypothesis,
nor even a fraction of the elegant work that went into developing tests of
the calcium hypothesis (for recent reviews, see Physiological Reviews, 1987~.
Rather, we would like simply to underscore the essence of the messenger
hypothesis: light-driven changes in its concentration were hypothesized to
cause gh,, to close rapidly.
Calcium as Internal Messenger
One line of evidence that strongly argued against Ca2+ playing the role
of internal messenger was developed by Lamb and colleagues (Matthews
et al., 1985; Lamb et al., 1986~. Using a combined suction electrode and
whole-cell gigaseal electrode (such as illustrated in Figure 2), they infused
the calcium buffer/chelator BAPTA into rods and demonstrated that the
OCR for page 63
PHOTOTRANSDUCTION IN YERTEBRATE RODS
63
FIGURE 2 A pair of photographs of a salamander rod whose outer segment is held in
a suction pipette and whose inner segment has been impaled with a gigaseal electrode; a
cone was fortuitously attached to the rod in the inner-segment region. The top picture
was taken with transmitted light, the bottom picture with epifluorescence. The cell was
intially drawn into the suction pipette under deep infrared illumination. It was then
recorded from for approximately 1 min after intracellular access was gained with the
gigaseal electrode (as in Figure 3), which contained, in addition to an intracellular medium,
5 mM 5~6~-carboxyfluorescein. The epifluorescence light was then turned on, and the
picture on the right was taken; finally, the transmission photograph was made. This
experiment demonstrates that substance infused into the inner segment penetrates to the
outer segment. The relative differences in intensity of fluorescence of the outer and inner
segments probably reflect the differences in the water space—in the outer segment more
than half of the volume is occupied by the disks.
OCR for page 64
64
E.N PUGH, JR, W.H. CORES, AND ID. TONER
initial phase of the photocurrent was not altered by the chelator (as required
by the calcium hypothesis), even though internal evidence was strong that
much of the chelator did not have calcium bound to it.
The Ca2+ messenger hypothesis was mortally wounded when Yau and
Nakatani (1984) showed that the outer-segment, light-sensitive membrane
current could be carried by pure Ca2+, and that even in the absence of
voltage control an inward calcium current of 300 pA, through the light-
sensitive conductance, could be sustained for several seconds. The 300 pA
of calcium current corresponds to an influx of about 109 Ca2+/second, yet
even so required about 5 see to completely shut down the light-sensitive
conductance. In contrast, a 200-msec flash delivering about 1000 isomer-
izations during the calcium influx shut down the current in about 100 msec.
A simple calculation shows that, to produce more calcium than has al-
ready come in through the dark current, the flash would have had to cause
the release of more than 106 C82+ ions per isomerization in about 200
msec, a much greater release stoichiometry than anyone ever postulated;
indeed, the maximal reliable releases by light ever obtained from disk mem-
brane suspensions were on the order of a few calcium ions per isomerized
rhodopsin! Yau and Nakatani's (1984) experiments also demonstrated the
existence of an Na/Ca exchange in the outer-segment plasma membrane,
a mechanism that uses the energy in the Na gradient to evict Ca2+. In
a subsequent experiment, Yau and Nakatani (1985) showed that about 5
percent of the normal dark current is carried by calcium moving together
with sodium; thus, Ca2+ in the outer segment actually declines during the
light response. This interpretation was directly confirmed by McNaughton
et al. (1986), who administered the coup to the hypothesis that an elevation
in internal calcium in the outer segment carries the message of excitation
between disk and plasma membrane.
Cyclic GMP as Internal Messenger
Bill Miller and Grant Nicol (Nicol and Miller, 1978; Miller and Nicol,
1979) were responsible for the beginnings of the electrophysiological testing
of the cGMP-transmitter hypothesis: they discovered that injection of
cGMP into outer segments depolarizes rods in a magnitude and with a
duration that is dose dependent and that light antagonizes the depolarizing
effect of cGMP. Miller (1982) extended the work with microelectrode
potentiometric recording and presented a series of detailed phenomena
consistent with the hypothesis that cGMP acts to open an outer-segment
conductance that is shut by light. Consistent with the Miller-Bitenskr
hypothesis, MacLeish et al. (1984) showed that cGMP infused into rods
induced an increased inward current that was reversed by light. At about
the same time, we (Cobbs and Pugh, 1985; Matthews et al., 1985) developed
OCR for page 65
PHOTOTRANSDUCTION IN VERTEBRATE RODS
Effect of cGMP on Rod Dark Current and Photocurrent
A
W
a' ~
,=~200i _
O a a a a a a a a b a a c a
65
1 1 1 _ 1 1 1 1 1 1 1 1 1 1 1 1 1 ~ 1 1 1 7~ 1 ~
0 20 40 60 80 100 120 140 160 180 300
Seconds
FIGURE 3 An experiment like that illustrated in Figure 2 in which the gigaseal pipette
contained 5 mM garlic GMP. The suction electrode records only outer-segment membrane
current. At AP an attached patch (having a resistance greater than a gigohm) was formed;
at WC the patch was reptured By slight negative pressure and whole-cell recording mode
commenced. A large inward current developed rapidly, which was entirely light suppressible.
At W the whole-cell electrode was withdrawn and the membrane patch resealed; over the
next 5 min the cell recovered completely to baseline. Positions marked with arrows indicate
20-msec flashes: a, 1000 isomenzations; b, 100; c, 10. The inset b is a tracing from the
infrared video record of the expenment. SOURCE: Cobbs and Pugh (1985~.
a combined suction electrode and intracellular electrode salamander rod
preparation like that used by Bavlor and Nunn fl9861, incorporating a
~ ~ ~ ~ ~ ~ . . ~
glgaseal electrode rather than a mlcroelectrode, as illustrated in blgure i.
Some of our results are shown in Figure 3.
Since the suction electrode in Figure 3 records only from the outer-
segment membrane, this experiment and similar ones by Matthews et al.
(1985) demonstrated that the current induced by cGMP is located in the
outer segment. In another set of experiments with the same technique
(Cobbs et al., 1985), it was demonstrated that cGMP has the same effect
on cones—that is, it increases an outer-segment inward current that light
suppresses.
The cyclic GMP messenger hypothesis came into full flower with the
exciting discovery by Fesenko et al. (1985) of a cGMP-gated cation conduc-
tance in excised patches of frog outer-segment membrane, as illustrated in
Figure 4.
Fesenko et al. presented several arguments for identifying the cGMP-
gated conductance with the light-sensitive conductance, including I/V rela-
tion, noise spectrum, and ion specificity, although all these fingerprints were
somewhat discrepant from what was expected. Soon thereafter, however,
identification of the cGMP-activated conductance, ROMP, with the light-
sensitive conductance, go, was made firm by Yau and Nakatani (1985) and
OCR for page 66
66
4 -
o
o
o ~
E
~ 4-
o
,`
._ ~
.~ an
CL
C:
Cal
+\
~ X
—\
o At
_W
O\+
O O
O — O
[(I - ~ )/i] 6°l
I
O? ~ C
0~/
in
~ (
_ U)
o o.
o
\
O \
X
O ' - ,
o ye`
o
O
O
Lo
O
.
Cam
C:
O ~
LO ~
_
O
O
O
O
E ~
=_?
o o ~
L' ~ 1
_1
m ~
~ s 0 ~ ~ ~
a Z ~ ~ ~ ~
~ O ~ ~ Cal 0~
o .= 8 c,
it ~ 0 (U
S ='>~= ~
~ c_
0 ~ 0
au ~ ~
c~ ~ ~ ~- 11 -
=l ~ %
o ~ ° :; 3
~ s ~ s
~ 3 ~ ~ ~ ~
_ ,,,
+ ,=
_, _
C) L~ ~ ~ h7.o
s ~ s ~ ~ 3
o ~ o ~ _
~ E ~ ~ ~ ~o
S ~ ,c =' ~ >
·_ Ct ~,) ~ ~ =~
—'~D ~ S .0 - ' ="
_
~ ~ ~ ~ ~ o
~ a C Zc = a ~
0 ~
ca 0 ~ _
a ~ ~ , , =~
m ,ps, _ a~ O
~˘ ~ ~ c., ~ ~
~ ~ ct ~ S c~
~ a ~ ;: C ~
11
c
o
c~
-
-
cO
- -
s' oo
~ - .
3 ~ ~o
OCR for page 67
PHOTOTRANSDUCTION IN VERTEBRATE RODS
67
by Matthews (1986, 1987~. Matthews, for example, demonstrated conclu-
sively that the light-suppressible channel noise recorded from cell-attached,
outer-segment membrane patches had a spectrum identical to that recorded
when the identical membrane patch was pulled and perfused with cGMP.
The cGMP Cascade: A Biochemical Sketch
Figure 5 shows a sketch of the so-called cGMP cascade of the outer
segment. This diagram summarizes not only the electrophysiological work
cursorily reviewed above but also a virtual flood of biochemistry that
followed in the wake of Bitensky et al.'s (1971) hypothesis (space does not
even permit even a cursory review of the relevant biochemical literature;
for extensive reviews, see Miller, 1981; Stryer, 1986; Pugh and Cobbs, 1986;
Liebman et al., 1987~.
According to the canonical version of the cascade theory, the sequence
of events is as follows: light does only one thing isomerize the rhodopsin
chromophore, causing it to undergo a conformation change that converts
it into an enzyme. As an enzyme, R* catalyzes the binding of GTP to a
protein, GTP-binding protein, which is present in an amount of about one
copy for every 10 rhodopsins. Several lines of evidence say that one R* In
the absence of the inactivation process can activate several hundred G-
proteins/second (Liebman and Pugh, 1979, 1982; Vuong et al., 1984~. Under
normal intracellular conditions, however, R* is inactivated fairly rapidly,
apparently by phosphorylation and the binding of "48K protein." G* is not
an enzyme but rather a general messenger with its own limited lifetime.
While active, G* in turn activates, by direct binding, the third protein in the
cascade, phosphodiesterase (PDE). PDE is a powerful enzyme capable of
catalyzing the breakdown or hydrolysis of about 2000 cGMPs per second.
One should keep in mind, however, when describing the "power" of PDE
that it is likely that the pee cGMP in the rod outer segment is fat lower
than the km of the enzyme, as guessed in 1985 (Cobbs and Pugh, 1985),
and that the hydrolysis rate is always scaled by the km. According to the
theory, PDE* next reduces the local free cGMP; as the cGMP unbinds
from the conductance, it is not replaced, causing ghu to close.
This canonical cGMP cascade theory requires fleshing out, even to be a
complete qualitative ("what does what") theory and, indeed, is well known
to contain several unresolved problems. Most of these problems have to do
with the kinetics of the restorative reactions, which are shown in the open
arrows in Figure 5. A key problem that remained unresolved after calcium
was rejected as the excitational messenger was the nature of its role, if any,
in transduction. This problem now appears near resolution. Having been
rejected as the excitational message, intracellular calcium concentration has
now made a major comeback as a restorative feedback signal and likely
OCR for page 68
68
4,
~n
D: ~ ~, a, ~ ~ c5 ~ x
a) 0
on ,~ i~
c ~ - ~
o ~ c c) E
_ o s ~ C ~ C
o s ~ C ·— ~ ~ c
tC C o ~ C ° C
t,, C S ~ ·~,, Q ) Q X
O Q ~ ~ O Y
S ~ CI) (D S Co ~ ~ Z
G)
C
~ _ ;;.,;.,; ;; ;;; ;:;:::::; :::::::::::::::::::::::::::::::::::::::::::::
E ~ ........ : : ..................... .
,,, ~ : . :.: :.:.:. :.:.:.: :.: :
CG ~
E
~ ~Y/9
~^.W ' .
~:
o
y
~r
A
-
D: ,
~ —
L
Ul
*,
11
: I _
1 ~
_ C]
+
,-
,
-
~!
,~
1~) 1
1
o 1
V,
o 1
o 1
,
mC?
1
1
.
z
~ I
c
a) 1,
o
~ 1
. 1 n5
\~
C)
+ .
, i+~m
,
-
A
Ir
~n
C~
Ct
-
a)
x
t.~3 Q
C)
C) — ~
=-- ~ ~
O ~ ~o --
Cc ~ ~ ,lC
a, >-
~ ° 3 — ~
~ C~ ..
C: ~ ~ ~ ~
Ct · _ ~
~ o o P ~
a, ~ ~
._
_
_ r,
. ~ ~ ·— ~
3 ~ ° c . C
~ Ct ~ _
. ~ ~ ~ ~
~ b ~ c D
t~ C;: ~ C-~ —
~ .= — C,0 .=
~ ~ `', t.~a
~,, _ c
°
~ ~ ._ ~
=. .= ~ ~ .o
~ =~_ ~
O ~ .=
3
·= Ct
._
_ ~ ~ ~ .
o 3 .<=
C ~ ~ C
~ C.~ ·—
[I) ~ Ct
~ -~= C ~ ~
o
-
OCR for page 69
PHOTOTRANSDUCTION IN VERTEBRATE RODS
R · R*
k1
G+GTP G*+D . D*
k3
69
5'GMP cGMP + (9hv)c ~ (NcGMP.9hv)o
- 5
FIGURE 6 A formal chemical schematic of the "activation" steps in the cascade. R*,
G*, and D* represent the activated forms of the proteins rhodopsin, G-protein, and
phosphodiesterase, respectively; gin, represents the cGMP-activated conductance, which is
opened by the cooperative binding of N = 2 to 3 cGMP molecules. A differential equation
representation of this reaction sequence is presented in Cobbs and Pugh (1987b).
as the internal adaptational message (Matthews et al., 1988; Nakatani and
Yau, 1988~. The primary way by which dynamic changes in internal Ca2+
affect the cascade appears to be via the guanylate cyclase reaction (Koch
and Stryer, 1988; Hodgkin and Nunn, 1988~. The figure reflects thinking
and experimentation up to about early 1989.
FORMAL THEORY
The remainder of this paper sketches our recent efforts to develop
a formal representation of the reactions that lead to the closure of ghU
(reactions represented by the filled arrows in Figure 5) and to test this
representation against responses of voltage-clamped rods to intense, brief
flashes isomerizing up to 20 percent of the rhodopsin. The strategy em-
ployed is diametrically opposite those pioneered by Penn and Hagins (1972)
and Baylor et al. (1974), both of whom developed formal analyses of the
dim flash or linear response and then attempted to incorporate appropriate
saturating nonlinearities to extend the analysis. Rather, we build the theory
to handle the responses to the most intense flashes first and then change
one critical parameter to see if it can explain the responses to less intense
flashes. A priori testing of a theory against the most nonlinear range of the
response of the rod seems a poor strategy. Nonetheless, we hope to show
that this strategy not only makes a great deal of sense but also actually
yields rich insight into the system.
Figure 6 shows a chemical representation of the relevant cascade
reactions. A differential equation representation of this chemical model
can be found in Cobbs and Pugh (1987~.
One of the interesting formal features of the cGMP theory is that
it relies on a "negative" messenger; that is, the signal or photocurrent is
produced by the messenger's removal. This feature is no doubt shared
by messengers in other systems but is nonetheless remarkable. Because
OCR for page 70
70
E.N PUGH, JR, W.~. CORES, AD ID. TONER
of this feature, the maximum speed of the photocurrent can in principle
be rate limited by any of several steps up to and including the rate of
unbinding of the cGMP from the conductance and its attendant closure.
In contrast, as pointed out by Penn and Hagins (1972), a transduction
theory that employs a positive messenger whose release rate is a monotonic
function of the number of Rotas produced by a brief flash predicts that the
photocurrent should not reach velocity saturation over the whole range of
bleach intensities. This absence of rate limitation is predicted because the
more intense the flash the faster the messenger should be released, and
thus the second-order binding reaction of the positive messenger to the
conductance (which closes it) should proceed at a speed ever increasing
with flash intensity.
Photocurrent Velocity and Delay Saturation
Figure 7 shows the voltage-clamp photocurrents of two rods stimulated
with a series of intense 21) psec flashes. Note that the velocity of the voltage-
clamp photocurrent appears to saturate at a relatively modest fractional
isomerization, about 0.001. By virtue of the voltage-clamp technique it can
be confidently asserted that this velocity limitation is not due to capacitative
loading the membrane potential could be shown to have changed less that
a millivolt during the response. Above 0.001 fractional isomerization, the
velocity-saturated photocurrent seems to translate laterally to the left with
increasing bleach fraction, and this translatory behavior itself saturates,
reaching its left-most position at about 0.1 fractional isomerization. This
latter saturation phenomenon we call delay saturation. The average appar-
ent latency of the velocity- and delay-saturated voltage-clamp photocurrent
is about 7 msec (Cobbs and Pugh, 1987~.
Theoretical Account of Velocity Saturation
Can a formal representation of the cGMP cascade theory account for
these phenomena, and is the account quantitatively consistent with the
biochemical data on the reactions? Theory suggests two ways in which the
approximately exponential velocity-saturated photocurrent could arise: one
is simply by virtue of the rate of cGMP unbinding from the channel and its
closure; the second is by virtue of the limitation posed by the maximal PDE
activity on the rate at which cGMP may be removed from the cytoplasm.
Employing numerical values taken from the biochemical literature Able
3 and 4 in Cobbs and Pugh, 1987) for most of the theory parameters,
we found that a velocity limitation imposed either by the intrinsic rate of
cGMP unbinding and channel closure (derived from noise power spectra)
of the conductance or by the maximal PDE activity (or a combination
OCR for page 71
PHOTOTRANSDUCTION IN VERTEBRATE RODS
Time from flash (msec)
A 0 10
,
20 30 40 50 60 70
I r I
1 1 1
a)
Q
Ct
-
a)
-
>
Q
a)
-
>
E
a)
-
~ ~-,~ C -
'l~1 o~3 V
B
0 10
1 1
\1,
~ o- 4
\
1 o- 5
~ ~ ;~ ~ ·— _
20 30 40 50 60 70
I I _ _ ~
-
p~ A —~r~ ~\ ~
~ '
W. ~
71
~ _
~ 10-45
v\
V:
~_~~ 10-55
Nit
~ _: ~ · it, ;~ = ~_~0
FIGURE 7 Voltage-clamp photocurrents of two rods stimulated with a series of intense
Parsec flashes, which isomers the fraction of rhodopsin given by the parameter near
each curve. In part A note that there is almost no change in photocurrent velocity for the
three most intense flashes, even though the left-most current was produced with a stimulus
1000 times more intense (10-1 vs. 10 4~: this is the phenomenon of velocity saturation.
In part A note that the 10-fold increase from 10 4 to 10 3 produces about the same
^~^ aria ~ ~ ~ a^`
late rat translation as the 100-fold increase from 10 3 to 10 1 and that in part B the
Refold change from 10 1 7 to 10-0 7 produces virtually no change in the lateral position
of the photocurrent: this is the phenomenon of delay saturation. Reprinted from Cobbs
and Pugh (1987) with permission.
OCR for page 72
72
E.N PUGH, JR, W.H. CORES, AND ~D. TONER
of both) produced solutions of the equations consistent with our data.
These two hypotheses about velocity saturation appear to be resolvable by
testing the theory against the photocurrents of rods infused with cGMP or
its weakly hodroyzable analog, 8-Bromo-cGMP (Zimmerman and Baylor,
1986; Barkdoll et al., 1988~. The extension of the theory must deal with
two serious complications: (1) the diffusion of the nucleotide is relatively
slow; (2) the rod outer segment has such a high longitudinal resistance (1
to 3 megohm/pm) that its cable properties must be taken into consideration
when modeling the cGMP-induced currents (Cameron and Pugh, 1988~.
Our work on extensions of the theory to incorporate diffusion and cable
properties is nearly complete, and we should know shortly which feature of
the cascade imposes the photocurrent velocity limitations.
Theoretical Account of Delay Saturation and Translation
Formal theory also provides interesting insight into the translatory
behavior that is seen at fractional isomerizations of about 0.001, and the
c. 7 msec saturation limit of the translation that occurs at about 0.1 fractional
isomerization. It is interesting that the delay saturation occurs at about
0.1 fractional isomerization, at which intensity there is one photolyzed
rhodopsin for every G-protein (the ratio of the two proteins being about
10/1~. One would thus expect delay saturation to set in at this flash intensity,
given that the time that it takes an isomerized rhodopsin to encounter a
G-protein is short. And indeed the time must be short; otherwise one
isomerized rhodopsin could not activate 500 to 1000 G-proteins/second
(Fung and Stryer, 1980; Liebman and Pugh, 1982~. Further insight can be
had by detailed comparison of theory and data.
Figure Sa compares average velocity- and delay-saturated photocur-
rents of a population of 11 rods (thicker, noisy trace) with a family of
theoretical curves (thinner traces). The left-most theoretical curve is a
particular parameterization of the theory, optimized to fit the data trace.
All the other theoretical curves are generated by changing a single pa-
rameter, the only rate constant in the theory expected to depend on light
intensity. This rate constant can be expressed as the pseudo-first-order rate
at which each rhodopsin activates the pool of G-protein to which it has
access before the entire G-protein pool is activated (at these flash levels
all the G-protein is predicted to become activated). To produce the series
of theoretical curves this one rate constant has been decremented in a
geometric series, each decrement approximately twofold. Clearly, the the-
ory provides a natural account of the translatory behavior. The underlying
cause of translation lies in the very nature of cascaded reactions followed
bv a rate-limiting step, not in any details of the parameters.
OCR for page 73
PHOTOTRANSDUCTION IN VERTEBRATE RODS
A
,:4 ~ ;~ ,,,
t ~ \ ~ ~
At
-
_ it\ _
\ \
\. ~
\ \
~ \.
. _ _
1 ~
30 40
0 10 20
Time from flash (msec)
B ~
~ 10-s
~ ~ An
fo4
t1o-3
10-1 ~ ~
I ~
0 10
i,
1 1 1
20 30 40
Time from flash (msec)
73
FIGURE 8 (A) The thickened noisy trace is the mean velocity- and delay-saturated
photocurrent of 11 rods stimulated with 20-ysec flashes that isomerized at least 2 percent of
the rhodopsin. The thinner traces are solutions of the differential equation representation of
the cascade equations in Figure 6, as given in Cobbs and Pugh (1987b), while the parameters
are those used in Figure 15A-C of the same paper. Only a single rate parameter, k20,
the effective rate with which each rhodopsin activates its respective pool of G-protein, was
varied to produce the theoretical traces. From far left to far right 1/k20 = 1.5, 6.0, 12.5,
25, 50, 100, 200, 400, 800, and 1600 msec. (B) Voltage-clamp photocurrents (noisy traces)
of a single rod stimulated with 20-psec flashes isomerizing the fraction rhodopsin indicated.
The smooth traces are generated by a representation of the cGMP-cascade theory: only one
parameter varies between the curves: k20, the pseudo-first-order rate with which rhodopsin
activates G-protein. From left to right, 1/k20 = 2.07, 167, 1225, and 13,038 msec.
OCR for page 74
74
E.N PUGH, JR, W.H. CORES, AND ID. TONER
An approximate functional dependence of the pseudo-first-order rate
of G-protein activation on light intensity can be derived from first-passage
theory (Szabo et al., 1980), if one assumes that the rate-limiting step in
activation of G-protein by rhodopsin is intramembrane lateral diffusion of
rhodopsin (Liebman and Pugh, 1981, 1982~; this in turn can be used to
predict the delay (fit in Figure &~. The agreement of predicted delay with
observation is quite good (see Cobbs and Pugh, 1987, Figure 12C). 1b test
the functional dependence in more detail, however, we need to be able to
estimate from photocurrents produced by different intensities on individual
cells the rate constant itself, not just the delay. Figure 8B shows how this
can be done. A family of responses from a single rod is simultaneously fit
with an instantiation of the theory by a parameter search routine; the only
parameter that is allowed to vary between curves is the rate constant in
question. Since we have independent knowledge of the fraction isomerized,
we can thus infer the dependence of the rate constant on isomerization.
We are in the process of confronting the theory in this fashion with the
individual records of many cells, in effect using the theory as a measuring
tool for estimating dependence of the rate on fractional isomerization and
thus testing the lateral diffusion hypothesis.
Efforts to build a formal representation of the cascade that could ac-
count for the voltage-clamp responses could not account for the magnitude
(7 msec) of the saturated delay unless one were to postulate that about 3 or
4 first-order delay steps of order 2 msec intervened between the binding of
rhodopsin to G-protein and the velocity-saturation mechanism (see Cobbs
and Pugh, 1987b, Appendix, for details). A natural mechanism for such
additional minor delay components would be in the binding of GTP, the
conformation change of the G-protein, the binding of G-protein to PDE,
and the latter conformational change as it becomes an active enzyme. Here
again, theory, applied to the photocurrents, allows us to "see" a biochemical
process that might otherwise escape attention.
SUMMARY
lithe hypothesis that there is an internal messenger in vertebrate pho-
totransduction has offered a rich challenge to scientists in many fields and
has brought together a wide array of tools and insights. With most of
the fundamental qualitative features of phototransduction now understood
(i.e., the identity of the molecular entities involved and the roles they play),
the field of phototransduction is entering a phase in which formal theo-
ries linking photoreceptor biochemistry and electrophysiology can play an
important role, permitting deeper understanding of this wonderful process.
OCR for page 75
PHOTOTRANSDUCTION IN VERTEBRATE RODS
REFERENCES
75
Bader, C.R., P.R. MacLeish, and E.A. Schwartz
1979 A voltage-damp study of the light response in solitary rods of the tiger
salamander. Journal of Physiology 296:1-26.
Barkdoll, NE., III, A. Sitaramayya, and E.N. Pugh, Jr.
1988 Kinetics of the hydrolysis of 8-Bromo-eydic GMP by the light-activated
phosphodiesterase of toad rods. Joumal of Neurochemistry 50:839 846.
Baylor, D.A, and B.J. Nunn
1986 Electrical properties of the light-sensitive conductance of salamander rods.
Journal of Physiology 371:115-145.
Baylor, D.N, AL. Hodgkin, and T.D. Lamb
1974 The electrical response of turtle cones to flashes and steps of light. Journal
of Physiology 242: 68~727.
Baylor, D.A., T.D. Lamb, and K.W. Yau
1979 The membrane current of single rod outer segments. Journal of Physiology
288:589~11.
Baylor, D.~, G. Matthews, and KW. Yau
1980 1\vo components of electrical dark noise in toad retinal rod outer segments.
Journal of Physiology 309:591~21.
Bitensky, M.W., R.E. Gorman, and W.H. Miller
1971 Adenyl cyelase as a link between photon capture and changes in membrane
permeability of frog photoreceptors. Proceedings of the National Acad~rny of
Sciences, USA, 68:561-562.
Bodoia, R.D., and P.B. Detwiler
1984 Pateh-elamp recordings of the light-sensitive dark noise in retinal rods from
the lizard and frog. Joumal of Physiology 367:18~216.
Cameron, D.A, and E.N. Pugh, Jr.
1988 Outer segment cable properties set upper limit to eGMP-induced clamp
current of salamander rods. Society for Neuroscience Abstracts, Pt. 1, p. 34,
#19.2.
Cobbs, W.H., and E.N. Pugh, Jr.
1985
1987
Eyelid GMP can increase rod outer segment light-sensitive current 10-fold
without delay of excitation. Nature 313:585-587.
Kinetics and components of the flash photoeurrent of isolated retinal rods of
the laIval salamander, Ambystoma Signum. Joumal of Physiology 394:52W
572.
Cobbs, W.H., NE. Barkdoll III, and E.N. Pugh, Jr.
1985 Cyclic GMP increases photocurrent and light-sensitivity of retinal cones.
Naa~cre 317:64 66.
Fesenko, E.E., S.S. Kolesnikov, and AL. Lyubarsky
1985 Induction by eyelie GMP of eationie eonduetanee in plasma membrane of
retinal rod outer segment. Nature 313:31~313.
Fung, B.B.K., and L. Stryer
1980 Photolyzed rhodopsin catalyzes the exchange of GTP for bound GDP in
retinal rod outer segments. Proceedings of the National Academy of Sciences,
USA, 77:2500 2504.
Hagins, W.A., R.D. Penn, and S. Yoshikami
1970 Dark current and photocurrent in retinal rods. Biophysical Joumal 10:380-
412.
OCR for page 76
76
Liebman, P.A.,
1987
E.N PUGH, JR., W.H. CORES, AND ID. TANNER
Hodgkin, A.L., and B.J. Nunn
1988 Control of light-sensitive current in salamander rods. Joumal of Physiology
403:439~71.
Koch, K.-W., and L. Stayer
1988 Highly cooperative feedback control of retinal rod guanylate cydase by
calcium ions. Nature 334:64 66.
Lamb, T.D.
1986 Transduction in vertebrate photoreceptors: the roles of cyclic GMP and
calcium. Trends in Neurasciences 9:22~228.
Lamb, T.D., P.A. McNaughton, and K.W. Yau
1981 Spatial spread of activation and background desensitization in toad rod
outer segments. Joumal of Physiology 319:463~96.
Lamb, T.D., H.R. Matthews, and V.R. Torre
1986 Incorporation of calcium buffer into salamander retinal rods: a rejection of
the calcium hypothesis of phototransduction. Joumal of Physiology 372:315
349.
Liebman, P.A., and E.N. Pugh, Jr.
1979 The control of phosphodiesterase in rod disk membranes: kinetics, possible
mechanisms and significance for vision. Vision Research 19:375-380.
1981 Control of rod disk membrane phosphodiesterase and a model for visual
transduction. Pp. 157-170 in Molecular Mechanisms of lingual Transduction,
W.H. Miller, ed. New York: Academic Press.
1982 Gain, speed and sensitivity of GTP binding vs. PDE activation in visual
excitation. Vision Research 22:1475-1480.
K.R. Parker, and E.A. Dratz
The molecular mechanism of visual excitation and its relation to the structure
and composition of the rod outer segment. Annual Review of Physiology
49:765-791.
MacLeish, P.R., A.E. Schwartz, and M. Tachibana
1984 Control of the generator current in solitary rods of the Ambystoma tig~inum
retina. Joumal of Physiology 348:645-664.
Matthews, G.
1986 Comparison of the light-sensitive and cyclic GMP sensitive conductances
of the rod photoreceptor noise characteristics. Joumal of Neuroscience
6:2521-2526.
1987 Single-channel recordings demonstrate that cGMP opens the light-sensitive
ion channel of the rod photoreceptor. Proceedings of the National Academy
of Sciences, USA, 84:58~585.
Matthews, H.R., V. Torre, and T. Lamb
1985 Effects on the photoresponse of calcium buffers and cyclic GMP incorporated
into the cytoplasm of retinal rods. Nature 313:58~585.
Matthews, H.R., R.LW. Murphy, G.L~ Fain, and T.D. Lamb
1988 Photoreceptor light adaptation is mediated by cytoplasmic calcium ooncen-
tration. Nature 334:67~9.
McNaughton, P.N, Lo Cervetto, and B.J. Nunn
1986 Measurement of the intracellular free calcium concentration in salamander
rods. Nature 322:261-263.
Miller, W.H., ed.
1981 Molecular Mechanisms of Photoreceptor Transduction. New York: Academic.
OCR for page 77
PHOTOTRANSDUCTION IN VERTEBRATE RODS
77
Miller, W.H.
1982 Physiological evidence that light-mediated decrease in cyclic GMP is an in-
termediary process in retinal rod transduction. Joumal of General Physiology
80:10~123.
Miller, W.H., and G.D. Nicol
1979 Evidence that cyclic GMP regulates membrane potential in retinal rod
photoreceptors. Nature 280:64 66.
Nakatani, K., and KW. Yau
1988 Calcium and light adaptation in retinal rods and cones. Nature 334:69-71.
Nicol, G.D., and W.H. Miller
1978 cGMP injected into retinal rod outer segments increases latency and am-
plitude of response to illumination. Proceedings of the National Acad~ny of
Sciences' USA, 75:5217-5220.
Penn, R.D., and W.N Hagins
1972 Kinetics of the photocurrent of retinal rods. Biophysical Joumal 12:1073-
1094.
Pugh, E.N., Jr., and J. Altman
1988 A role for calcium in adaptation. Nature 334:16-17.
Pugh, E.N., Jr., and WH. Cobbs
1986 Visual transduction in vertebrate rods and cones: a tale of two transmitters,
calcium and cyclic GMP. Vision Research 26: 161~1643.
Pugh, E.N., Jr., and ~J. Miller
1987 Phototransduction in vertebrates. Annual Review of Physiology 49:711-714.
Stirling, C.E., and ~ Lee
1980 [3H]oubain autoradiology of frog retina. Journal of Cell Biology 85:31~324.
Stryer, L.
1986 Cyclic GMP cascade theory of vision. Annual Review of Neuroscience
9:87-119.
Szabo, A., K Schulten, and Z. Schulten
1980 Fint passage time approach to diffusion-controlled reactions. Joumal of
Chemical Physics 72:435~4357.
Vuong, T.M., M. Chabre, and L Stryer
1984 Millisecond activation of transduction in the cyclic nucleotide cascade of
vision. Nature 311:659-661.
Yau, K.W., and K. Nakatani
1984 Electrogenic Na-Ca exchange in retinal rod outer segment. Nature 311:661-
663.
Light suppressible, cyclic GMP-sensitive conductance in the plasma mem-
brane of a truncated rod outer segment. Nature 317:252-255.
Yau, K.W., L.W. Haynes, and K. Nakatani
1986 Roles of calcium and cyclic GMP in visual transduction. In Membrane
Control of Cellular Activity, H.C. Luttgau, ed. Stuttgart: Gustav Fischer.
Yoshikami, S., and W.N Hagins
1970 Ionic basis of dark current and photocurrent of retinal rods. Biop;hysical
Joumal 10:60a.
Zimmerman, A.L., and D.N Baylor
1986 Cyclic GMP-sensitive conductance of retinal rods consists of aqueous pores.
Nature 321:7~72.
1985
Representative terms from entire chapter:
rod outer
