BBS SPECIAL ISSUE: Controversies in Neuroscience III: Signal Transduction in the Retina and Brain

WHAT ARE THE MECHANISMS OF PHOTORECEPTOR ADAPTATION?

Keywords

Abstract

Photoreceptor outer segments are a favorable preparation for studying adaptation processes.

Vertebrate photoreceptors show striking powers of adaptation, adjusting their gain to remain responsive to light transients as ambient light intensity increases over 3-5 log units. A primary locus of these adaptation processes is the outer segment portion of the photoreceptor, on which this article focuses. It is clear that these structures determine the basic parameters that define subsequent stages of visual information processing in the retina (cf. Shapley and Enroth-Cugell, 1984). The enzymes that regulate excitation and adaptation have been characterized mainly by studies on cattle and frog rod outer segments (ROS), and are described in several recent reviews (Chabre and Deterre, 1989; Detwiler and Gray-Keller, 1992; Hargrave and McDowell,1992; Hurley, 1992; Koutalos and Yau, 1993; Lagnado and Baylor, 1992; Pfister et al., 1993; Pugh and Lamb, 1990; Stryer, 1991; Yarfitz and Hurley, 1994; Yau, 1994). The milligram to gram quantities of ROS necessary for purification and analysis of transduction enzymes are more easily obtained from cattle retinas. Larger amphibian ROS, while more difficult to prepare in quantity, have the advantage that pure and physiologically active suspensions can be obtained. This permits study of both the electrophysiological and biochemical correlates of adaptation. The composition of these structures is better characterized than that of any other primary sensory organelle: The mass is half protein and half lipid (Fliesler and Anderson, 1983), 70% of the protein mass is rhodopsin, 17% is the G-protein transducin (Gt), and most of the balance is accounted for by proteins involved in cyclic nucleotide metabolism and protein phosphorylations (Hamm and Bownds, 1986). Rhodopsin is composed of the protein opsin covalently linked to 11-cis retinal, and the photochemistry of this complex has been well studied. Thus the system allows precise quantitation of input (the number of rhodopsin molecules hit by light). Sophisticated modelling and kinetic analysis not yet achieved in other internal messenger systems is possible (Bownds and Thomson, 1988; Sneyd and Tranchina, 1989; Forti et al., 1989; Dawis, 1991; Lamb and Pugh, 1992). An outline of how excitation works is largely in place, and interest focuses now on adaptation. What must be explained? Excitation and Adaptation.

A single photon hitting one of the 3x109 rhodopsin molecules present in disk membrane system of a dark frog ROS causes the closing of several hundred channels in the plasma membrane. This transiently halts the inward movement of several million ions. In the frog the duration of the dark adapted response is approximately four seconds with a time to peak of 900 msec (Baylor et al., 1979a). This excitation process is very stereotyped and reliable (Lagnado and Baylor, 1992). If a step of background illumination is turned on, the response relaxes to a plateau level within seconds after an initial peak. This process is usually called background adaptation. Superimposed responses to test flashes become smaller and more rapid as this background light is increased over 3-5 log units. At high backgrounds, the amphibian rod response peaks at approximately 300 msec and turns off 2 to 3 times faster (Fain, 1976; Baylor et al., 1980; Nicol and Bownds, 1989). The flash sensitivity (Sf), defined as the change in current per photon absorbed declines in a linear fashion as a function of the log of the background intensity. Until recently it had been supposed that background adaptation was a characteristic of the rods of lower vertebrates, and not displayed by mammalian rods. Yau and his collaborators, however, have now shown that adaptation behavior is observed in several warm blooded animals, including rats, rabbits, cows, and monkeys (Nakatani et al., 1991; Tamura et al., 1991).

If a significant amount of rhodopsin is bleached (greater than ~5%), the photoreceptor's sensitivity is at first abolished. In a subsequent dark period sensitivity begins to recover as opsin regenerates to form rhodopsin but remains low (bleaching adaptation) as long as opsin is still present (cf. Kahlert et al., 1990). Recent work suggests that the underlying mechanism may be similar to that of background adaptation (Clack and Pepperberg, 1982; Cornwall and Fain, 1992). The major differences between bleaching and background adaptation may be that in the former the large amount of opsin present has residual excitatory activity and the efficiency with which the ROS captures photons is lowered because less rhodopsin is present. The residual excitatory activity can be inhibited by adding retinal or some of its analogs (Jin et al., 1993). Fain and Lisman (1993) have suggested that residual excitatory activity may underlie photoreceptor degeneration associated with vitamin A deficiency or rhodopsin mutations that are constitutively active (Robinson et al, 1992).

The basic excitation pathway. Other articles in this review series have described reactions of the excitation cascade:

They are a variation on the ubiquitous pattern seen in many other G-protein systems. Recent work has shown that olfactory receptors use an analogous system, except that cyclic AMP and/or triphosphoinositol are the relevant channel regulators (Firestein, 1991; Breer and Boekhoff, 1992; Ronnett and Snyder, 1992). The interaction of excited rhodopsin (Rh*) with many G protein molecules (Gt, transducin), causes each to release GDP and bind GTP. Gt-GTP then activates cGMP phosphodiesterase (PDE) which hydrolyzes cGMP to 5'-GMP. The drop in cGMP caused by PDE activation causes closing of channels held open by cGMP in the darkness, halting a continuous entry of Na+ and Ca++ ions. The system inactivates as Rh* is quenched (see below) and the GTP of Gt-GTP is hydrolyzed to GDP, stopping PDE activation. The decrease in Ca++ entry activates a guanylate cyclase that enhances cGMP recovery. The excitation process thus increases GDP and 5'-GMP concentrations, and decreases GTP, cGMP, Na+, and Ca++ concentrations. All of these products are putative feedback regulators. Most interest, however, has focused on Ca++ and more recently cGMP. Because the rod metabolism efficiently buffers high energy phosphates (Groskoph et al., 1992) levels of GTP and ATP are maintained well above the binding constants of the reactions that utilize them. Thus variations in their concentration occur only at very bright intensities and are unlikely to be regulatory in background adaptation (Biernbaum and Bownds, 1985). Because we think it plausible that adaptation might occur at almost any the steps of the photoreceptor enzymatic cascade this discussion outlines the activation and inactivation of the primary reactions, considering each as a possible locus for adaptational controls. We will emphasize, where possible, measurements made on more intact and concentrated ROS preparation at the low light levels (between 1 and 105 Rh*/ROS/sec) at which rods normally function, and discuss technical problems that arise. It is important to point out that the majority of biochemical studies have used very unphysiological conditions: diluted suspensions of disrupted ROS (or purified enzymes) and illumination bleaching a substantial fraction of the rhodopsin present.

Excited rhodopsin catalyzes the activation of a G-protein, (Gt, transducin)

It is instructive to visualize rhodopsin excitation and subsequent transduction reactions with respect to the surfaces of single disk membranes. A typical amphibian ROS contains a stack of approximately 2,000 disk membranes that arise as evaginations of the ciliary plasma membrane during outer segment morphogenesis (Steinberg et al., 1980; Williams et al., 1988) and then become self enclosed (thus there are 4,000 disk membrane surfaces). The structure is continuously renewed throughout its lifetime. The disks occupy ~50% of the volume of the ROS, the balance is aqueous (Korenbrot et al., 1973). The surrounding plasma membrane contains the channels that ultimately are gated as a consequence of bleaching rhodopsin in the disk membranes. The ROS contains approximately 3 x 109 rhodopsin molecules (Liebman and Entine, 1968), or roughly 106 rhodopsins/disk membrane surface. This rhodopsin accounts for 15-20% of the disk surface, with individual rhodopsin molecules being approximately 2 nm apart, and colliding every 1-10 msec. Several recent reviews provide relevant detailed information about rhodopsin structure (Khorana, 1992; Nathans, 1992). The concentrations of rhodopsin, Gt, and PDE in the ROS are approximately 6,000, 600, and 22 mM, respectively, calculated with respect to the aqueous volume of the ROS (Hamm and Bownds, 1986; Dumke et al., 1994). The drawing below visualizes the relative abundance of the components. The square contains 1000 rhodopsin molecules (the smallest dots), 100 Gt (grey dots), 4 PDE molecules (large dots) and one free cGMP molecule (the small black dot shown between disk and plasma membrane). Phospholipids of the bilayer, ~50 for each rhodopsin molecule, are highly unsaturated and provide a fluid environment for lateral interactions of the protein components (Fliesler and Anderson, 1983). The number of channels (the large dot in the plasma membrane) is approximately 1 for every 10 of these frames.

When one of the rhodopsin molecules in a ROS absorbs a photon, there is 50% probability that isomerization of the 11-cis retinal chromophore to the all-trans configuration will occur, and lead to the formation, on a millisecond time scale, of metarhodopsin II (abbreviated as Rh*) and generation of the photoresponse (Baylor et al., 1979b). Photometric measurement of the absorption shift caused by photon absorption can yield the number of rhodopsin molecules that form the metarhodopsin II intermediate that interacts directly with Gt. In no other system can receptor activation be so accurately specified. Knowing the number of excited rhodopsin molecules permits us to calculate the gains of subsequent steps in the excitation pathway with respect to the initial input. We can then ask whether this gain changes during adaptation processes.

It is unlikely that the initial excitation step is a site of adaptation. This would be the equivalent of saying that the quantum efficiency of bleaching changes during adaptation, and electrophysiological measurements suggest that the probability that photon absorption will generate a photoresponse is not altered by dim background illumination (Baylor et al., 1979b). It is possible, however, that control of the lifetime of metarhodopsin II (Rh*) could be a factor in adaptation and this will be considered below with the reactions that terminate its activation.

Excited rhodopsin exposes within milliseconds a protein surface that binds to the Gt-GDP, permitting the exchange of GTP for GDP on the a subunit. The crystal structures of both the GTP- and GDP-bound forms of the a subunit recently have been determined (Noel et al., 1993; Lambright et al., 1994). Gta-GTP (Gt*) is released from Rh*, accompanied by dissociation of the b- g subunit complex, and the process repeats many times:

Interaction between Rh* (an intrinsic membrane protein) and Gt occurs via translational diffusion and collision on the disk surface (which contains 106 rhodopsin and 105 Gt molecules, Hamm and Bownds, 1986). Because very little Gt-GTP formation occurs in the dark, radioactive GTP-a-32P or GTP-g-35S can be used to monitor its' generation over physiological ranges of illumination that bleach 102 to 105 rhodopsin molecules/s. High gains (~40,000 Gt*/Rh*) have been observed in electropermeabilized ROS that retain most of their complement of Gt (Gray-Keller et al., 1990). In this preparation rates of 200-400 Gt*/Rh*/sec have been obtained, while measurements of the light scattering changes which accompany Gt activation have been interpreted to indicate rates as high as 1000 Gt*/Rh*/sec (Vuong et al., 1984; Uhl, 1990). This discrepancy needs to be resolved. A curious feature is that Gt-GTP can be generated not only by rhodopsin bleaching but also by mechanical disruption of ROS (Gray-Keller et al., 1990), or freezing and thawing (Klenchin and Bownds, unpublished data). The mechanism for this activation is not clear.

Might the rate of Gt activation be altered by adaptation chemistry? >From a functional perspective it would make little sense at moderate background levels of light to have the onset of the photoresponse slowed by adaptation, for a change in light levels needs to be sensed just a rapidly as the response to the initial onset of illumination. What is relevant in the presence of background light is that the amplitude of the response per absorbed photon be smaller and that it terminate more rapidly. Thus one would expect mechanisms that determine inactivation times are more likely loci of adaptational control than those that determine initial kinetics.

Two lines of evidence suggest that the initial kinetics of Gt activation may not be sensitive to the previous history of illumination. Kahlert et al. (1990) have shown that the presence of bleached rhodopsin does not alter the sensitivity or gain of a light-scattering signal recorded from a living bovine retina which is taken to monitor Gt activation. For reasons that are not understood, similar signals can not be recorded from amphibian retinas. Earlier reports also indicated that background illumination has very little effect on the initial kinetics of a superimposed flash response (Baylor and Hodgkin, 1974; Lamb,1984). However, more recently Lagnado and Baylor (1994) have shown in recordings from truncated amphibian rods that reproducing the Ca++ concentration decrease that occurs during background illumination lowers the initial slope of the flash response. Direct chemical measurements of the initial kinetics of Gt* production after a dim flash, and then after a dim flash superimposed on background illumination, might resolve the issue.

Gt, as well as PDE and several other transduction enzymes, is covalently modified by lipid attachment. Farnesylation of the g subunit of Gt enhances its activation, as does further methylation (Fukada et al., 1990; Ohguro et al., 1991; Perez-Sala et al., 1991). Neubert et al. (1992) have characterized four different N-terminal acylations of Gta. Anant et al. (1992) have demonstrated differential prenylation in vivo of the a and b subunits of PDE by farnesylation and geranylgeranylation, respectively. They also report in vivo farnesylation of rhodopsin kinase (Anant and Fung, 1992). While it has been suggested that these reactions might regulate enzyme activity, it seems unlikely that they act on the time scale of excitation and adaptation. They are more likely to be long term structural modifications whose purpose is to enhance membrane-protein association so that reactants are favorably poised to interact with each other in the two dimensional surface of the disk membrane.

The Gt cycle completes when Gt*-GTP converts to Gta-GDP (see below) and reassociates with Gbg. Phosphoproteins are found in both frog ROS (Components I and II; Polans et al., 1979; Suh and Hamm, 1988; Hamm, 1990) and bovine ROS (Phosducin, Lee et al., 1987, 1990a) that are dephosphorylated after bright illumination and form complexes with Gbg. If this complex formation competes with reformation of Gtabg, a possible adaptation mechanism is suggested: depletion of the pool of Gtabg because its re-assembly has been blocked. The phosphoproteins are present in an amount comparable with Gt. However, their subcellular distribution is not clear. Large amounts are found in the inner segment portions of rod receptor cells (Lee et al, 1988; 1990b). Does adaptation modulate the duration of Gt activation?

This brings us to consider the processes that terminate the generation of Gt*. Reversal of the PDE activation caused by Gt* in vitro is greatly delayed by the removal of ATP (Liebman and Pugh, 1979, 1980), and the photocurrent in whole cell clamp recordings (Sather and Detwiler, 1987) or truncated ROS (Nakatani and Yau, 1988a) approaches physiological time scales (seconds) only if ATP is present. It is commonly assumed that the main function of ATP is to quench Gt* generation by serving as substrate in the phosphorylation of metarhodopsin II by rhodopsin kinase:

This light-dependent reaction leads to incorporation of up to 9 phosphates per Rh* molecule (Wilden and Kuhn, 1982). The idea that the effect of ATP is mediated by Rh* phosphorylation is confirmed by two independent lines of evidence. First, the effect of ATP can be observed only in the presence of rhodopsin kinase (Sitaramayya and Liebman, 1983a; Sitaramayya, 1986). Second, the ability of phosphorylated Rh* to activate Gt is reduced (Arshavsky et al., 1985; Wilden at al.,1986; Miller et al., 1986). Arshavsky et al. (1987) have demonstrated that the rate of Gt activation decreases 1.5 to 3.3-fold as phosphate incorporation increases from 2 to 6 phosphates/rhodopsin, and that the inhibitory mechanism involves an increase in the time necessary for the activation of each bound Gt rather than a reduction in the binding affinity of Gt for phosphorylated Rh*. Rhodopsin phosphorylation is followed by binding of a 48kD protein, arrestin, a "capping" reaction that blocks access of Gt so that no further excitation can occur (Wilden et al., 1986). Palczewski et al. (1992) have demonstrated that sangivamycin (an inhibitor of rhodopsin kinase) and phytic acid (an inhibitor of arrestin binding to phosphorylated rhodopsin) slow recovery of the photoresponse when they are dialyzed into Gecko ROS. Interestingly, they show also that high concentrations of sangivamycin cause changes in the light response that cannot be explained by selective inhibition of rhodopsin kinase, and suggest that other protein kinases, for example protein kinase C, may be needed for normal rod function. Hofmann et al. (1992) have recently shown that recycling of rhodopsin back to the active pool requires reduction of the retinal chromophore to retinol. Only then does arrestin dissociate, opsin becomes dephosphorylated, and opsin combines with 11-cis retinal to form rhodopsin again.

The shut off of rhodopsin activation must be very reliable and irreversible, so that only one burst of activity is set off by each photon absorption. Lagnado and Baylor (1992) make the point that quenching of rhodopsin by binding of arrestin to rhodopsin with one or two phosphates would not explain the constancy of the single photon responses observed. They suggest that reproducibility of the single photon response might be due in part to feedback control of rhodopsin's active lifetime.

Can rhodopsin phosphorylation be measured in fact to occur before, or on the same time scale as, the turning off of the photoresponse? A positive answer to this question was obtained by Sitaramayya and Liebman (1983b) demonstrating in bovine ROS that bleaching approximately one rhodopsin in each photoreceptor disk causes incorporation of about 20 phosphates per bleached rhodopsin in the first two seconds after the flash. Since each Rh* can not incorporate more than nine phosphate groups, this suggests that not only Rh* but also some non-activated rhodopsin can be phosphorylated, confirming the initial report of Bownds et al. (1972). A slower rate of Rh* phosphorylation, about one phosphate per bleached rhodopsin in the first 1-2 seconds after a light flash, was documented for frog ROS by Binder et al. (1990). This lower stoichiometry does not necessarily reflect a difference between the two species because higher bleaching levels of about 1000 Rh* per disk were used in the frog study. It would appear then that incorporation of one or two phosphates per mole of rhodopsin is adequate for photoresponse quenching. Experiments of Bennett and Sitaramayya (1988) suggest that only 1 or 2 phosphates/Rh* may be sufficient to promote arrestin binding and maximally quench the interaction between rhodopsin and Gt. The residues initially phosphorylated have recently been identified as serines 338 and 343 and threonine 336 (Papac et al., 1993; McDowell et al., 1993; Ohguro et al., 1993). The data are consistent with a role for rhodopsin phosphorylation in quenching Gt* generation, but a determination of the kinetic correlations between the time course of rhodopsin phosphorylation and the decay of Gt* generation has not been accomplished. An alternative view is that the crucial event in quenching Rh* is the binding of kinase, rather than the subsequent phosphorylation that is catalyzed by the kinase (Pulvermuller et al., 1993). In this model the phosphorylation would serve the function of insuring that the initial quenching was made permanent. However, the fact that ATP is required for quenching the cascade means that kinase binding by itself does not insure Rh* inactivation.

Is modulation of rhodopsin phosphorylation a possible site of adaptation? Lagnado and Baylor (1992) suggest that calcium might influence Rh* shutoff, making it more stereotyped because it would be determined by the fall in calcium concentration that follows channel closure (see below). Further, increasing the rate of Rh* phosphorylation during background illumination could potentially cause acceleration of photoresponse recovery of the sort observed during background adaptation. Recently Kawamura and his collaborators have provided convincing evidence that just such a scenario may be appropriate, by showing that decreasing Ca++ concentration shortens the duration of the photoresponse (Kawamura and Murakami, 1991). (It should be noted that the physiological data on Ca++ regulation of Rh* lifetime has thus far been obtained in broken truncated ROS whose interior is exposed to a bathing medium.) Analysis of photoresponse recovery from bright flashes in intact cells leads Pepperberg et al (1992, 1994) at argue that this lifetime does not change during adaptation, but rather that a later stage is involved.

Kawamura's group has identified a protein, named S-modulin (M.W. 23 kDa), that confers Ca++ sensitivity on rhodopsin phosphorylation (Kawamura et al., 1992; Kawamura, 1993). This protein turns out to be the frog homolog of bovine recoverin (Dizhoor et al., 1991), a Ca++ binding protein originally proposed to regulate guanylate cyclase (see below). The three-dimensional structure of recoverin recently has been determined (Flaherty et al, 1993). Gray-Keller et al. (1993) have shown that internal perfusion of Gecko rods with Gecko S-modulin as well as recoverin delays the recovery of the photoresponse. This correlates with the observation that at high Ca++ concentration these proteins inhibit rhodopsin phosphorylation, perhaps by acting on rhodopsin kinase. The inhibition is relieved if Ca++ is lowered to the level caused by illumination (< 100 nM). This suggests that the lowering of Ca++ that occurs during background adaptation should decrease the lifetime of Rh* and thus the duration of PDE activation, causing smaller and faster responses to superimposed flashes. Both earlier (Robinson et al., 1980; Kawamura and Bownds, 1981) and more recent (Kawamura and Murakami, 1991) experiments had shown in fact such an effect of Ca++ on PDE activation. The data of Wagner et al. (1989) also suggest Ca++ regulation of rhodopsin phosphorylation. A light scattering transient that they interpret as rhodopsin deactivation in ROS suspensions is accelerated as calcium is lowered from 1 to 0.1 mM. Recoverin appears to be essential for the Ca++ regulation, for Palczewski et al. (1988a) have found that Ca++ does not act directly on rhodopsin kinase. Most recently Chen and Hurley (1994) have reported that recoverin binds and imparts Ca++ sensitivity to rhodopsin kinase.

The situation is not as clear with respect to cGMP as a putative feedback regulator of rhodopsin phosphorylation. Hermolin et al. (1982) and Schuster and Farber (1984) noted inhibition of rhodopsin phosphorylation by cGMP, but more recently Palczewski et al. (1988b) and Binder et al. (1989) found no effect.

Another possible role for Rh* phosphorylation in light-adaptation can be suggested: If a substantial portion of non-bleached rhodopsin were to become phosphorylated during dim background illumination, its excitation should result in photoresponses with lowered amplitude and accelerated recovery. The amplitude reduction would be expected from the reduced ability of phosphorylated Rh* to activate Gt, while turnoff acceleration would result from faster arrestin binding to pre-phosphorylated rhodopsin. A hint that this might take place has come from measurements of rhodopsin phosphorylation in electropermeabilized amphibian ROS. At light levels in the operating range of the rod, bleaching one rhodopsin molecule can result in the incorporation of phosphate groups into several hundred dark rhodopsin molecules (Binder et al., 1990.). The gain diminishes rapidly as each interdiskal space begins to receive more than one photon, and the high gain reaction is lost if outer segments are fragmented into smaller pieces. This high gain reaction is sensitive to calcium, being more active at 10 nM than at mM calcium levels (Calvert and Bownds, unpublished observations). In permeabilized ROS, however, the high-gain phosphorylation reaction saturates when it has acted on less than 1% of the total rhodopsin present. This would not significantly alter the probability of an ROS generating a response upon photon absorption. It is possible, however, that measurements on the living retina might show that the high-gain rhodopsin phosphorylation reaction alters a more significant fraction of the visual pigment present. Such measurements have not yet been done.

The mechanism of the high gain phosphorylation remains to be determined. Recent studies (Fowles et al., 1988; Palczewski et al., 1991;Buczylko et al., 1991) have shown that rhodopsin kinase, on binding to bleached rhodopsin, becomes able to phosphorylate rhodopsin C-terminal peptide fragments. Perhaps it also phosphorylates the C-terminus of some dark rhodopsin molecules. An alternative possibility is that another kinase acts on rhodopsin. Newton and Williams (1991) suggest a possible analogy with the b-adrenergic receptor system, where both cAMP dependent kinase and b-adrenergic receptor kinase phosphorylate the receptor (Roth et al., 1991), with the cAMP dependent kinase more important at low receptor occupancy and the b-adrenergic receptor kinase acting at higher levels of receptor occupancy. The observation (Newton and Williams, 1993) that rhodopsin is the major substrate of protein kinase C in ROS raises the possibility that this enzyme may play a role corresponding to the cAMP dependent kinase of the b-adrenergic receptor system. Another possible mechanism leading to the accumulation of non-bleached phosphorylated rhodopsin in photoreceptor cells is an inhibition of the dephosphorylation reaction without blocking the regeneration of bleached rhodopsin by 11-cis-retinal. This possibility is discussed by Biernbaum et al. (1991) who demonstrated that a dim background illumination prevents rhodopsin dephosphorylation in a preparation of intact frog rods. Excited Gt activates PDE. The next step in the phototransduction cascade is the action of Gt* on its effector, PDE. The PDE holoenzyme is an abg2 tetramer. The a and b subunits each contain one catalytic and one, or possibly two, non-catalytic cGMP binding sites (Gillespie and Beavo, 1988, 1989; Charbonneau et al., 1990; Li et al., 1990; Cote and Brunnock, 1993). The two identical g subunits serve as inhibitors of the enzyme (Hurley and Stryer, 1982; Deterre et al., 1988; Fung et al., 1990). Full activation requires the binding of one Gt* molecule to each of the two inhibitory g subunits. Each frog disk face contains ~3,700 PDE molecules, anchored by prenylation to the membrane surface. An excited Gt* contacts one of these PDE molecules rapidly enough to activate it within milliseconds (Heck and Hofmann, 1993). This reaction is occurring by lateral diffusion on the surface of the disk. Given that the binding constant between PDE and Gt* is approximately 100-600 nM (Bennett and Clerc, 1989), lower than the concentration of Gt* at the disk surface (which, from above, is >1 mM), essentially all of the Gt* has activated PDE within milliseconds. [In diluted fragmented ROS, a smaller fraction of Gt* is successful in activating PDE (Liebman et al, 1987)]. Pugh and Lamb (1993) point out that each of the activation reactions in the transduction pathway can be treated as a short first order delay stage, so that the time course of PDE activation is a delayed ramp, with slope proportional to light intensity; the initial delay is about 10-20 ms (Lamb and Pugh, 1992).

The PDE activation-inactivation sequence occurs in a fundamentally different context than the Rh* - Gt sequence. In the latter case the reactants are poised for reaction but silent in the dark ROS, awaiting the discharge induced by illumination. PDE activation, on the other hand, modulates an ongoing flux: Both biochemical and electrophysiological studies (Dawis et al., 1988; Hodgkin and Nunn, 1988) suggest flux rates on the order of one turnover of the entire pool of unbound cGMP per second, and Pugh and Lamb (1990) estimate dark levels of free cytoplasmic cGMP to be 4 mM. This means that each interdiskal space has about 650 molecules of cGMP, all turning over once a second. Direct measurement of the dark PDE activity that underlies this flux is complicated by the dilution of reactants that occurs when ROS are disrupted to assay the enzyme. This disruption causes some dissociation of the inhibitory g subunit of PDE and thus increases dark background activity (cf. Arshavsky et al., 1992). Measurements performed under conditions preventing PDEg loss from PDE catalytic subunits indicates that the PDE activity in the dark is at least 300-fold less than the activity of light-activated enzyme (Arshavsky et al., 1992). Taking this fraction of the light activity, the cGMP concentration in the dark as constant and equal to 4 mM, PDE concentration as ~22 mM (see above), Vmax (in the light) ~4700 s-1 and Km as ~95 mM (Bownds et al., 1992; Dumke et al., 1994) Michaelis-Menten calculations yield a dark flux estimate for cGMP of less than 15 mM/sec. This number is reasonably close to the 4 mM/sec estimated in vivo (Pugh and Lamb, 1990). The effect of the large increase in PDE activity caused by light can be visualized by making a simple calculation of the amount of cGMP hydrolyzed after absorption of a single photon (i.e. 1 Rh*) in a frog rod. More precise calculations are reported in Lamb and Pugh (1992). Let us suppose that the rate of Gt activation is 1000 s-1 and that this rate is constant during the time of the photoresponse. Therefore, an average of 500 Gt* molecules are present during the rising phase of the response (~1 sec). Since the complete activation of a PDE molecule demands two Gt*, the average number of PDE molecules activated during this time will be about 250 molecules. Taking Vmax of activated PDE as about 4700 turnovers/sec and Km as 95 mM (Dumke et al, 1994) we can calculate that each activated PDE will hydrolyze about 200 cGMP molecules. The total amount of hydrolyzed cGMP then will be on the order of 50,000 molecules. This corresponds to the total cGMP content of approximately 80 interdiskal spaces or 4% of the whole pool of unbound rod cGMP. This is reasonably close to the length of ROS whose conductance is suppressed following a single photon absorption (see below).

It is likely that a considerably smaller change in cGMP would be sufficient to suppress this conductance. Because gating of the channel by cGMP is cooperative, hydrolysis of only a fraction of the cGMP present might be sufficient to cause a large conductance change. Overall, it would seem more efficient for cGMP to drop by a small amount over a longer length of the ROS than to undergo a large change over a small length (cf. Lamb and Pugh, 1992; Rispoli et al, 1993). Pugh and Lamb (1993) have pointed out that these kinetic parameters can generate very different relative changes in cGMP concentration if ROS volume is decreased, as in mammalian rods and cones. Equivalent cascade activation in a smaller ROS causes a larger fractional change in the cGMP pool.

The PDE enzyme is so active that it depletes cGMP in the interdiskal space very rapidly. As a result, the diffusion of cGMP into the interior of the disk stacks can become rate limiting in conventional biochemical measurement of cGMP hydrolysis in ROS fragments. This can have the effect of raising the apparent Km of PDE as well as limiting its apparent maximal velocity. Recent studies show that true values for Km and kcat of PDE (~95 mM and 4700 sec-1) are approached only when ROS membranes are completely disrupted into small vesicles (Dumke et al., 1994). Techniques for directly measuring these parameters in intact cells are not currently available. Perhaps the most accurate estimates of PDE activation kinetics in intact ROS come from reverse calculations based on the kinetics of the photoresponse (cf. Hodgkin and Nunn, 1988; Pugh and Lamb, 1993). The turnoff of activated PDE is a putative point of light adaptation.

The action of Gt*, like that of other G-proteins, has been thought to be terminated as its bound GTP is hydrolyzed by an "intrinsic GTPase activity" (cf Bourne et al., 1990, 1991). This activity measured in dilute suspensions has been much too slow (Fung et al., 1981; Baehr et al., 1982) to explain the rapid turnoff of PDE in suspensions of bovine ROS (Sitaramayya and Liebman, 1983b). More recent work has indicated that transducin GTPase under more physiological conditions is much faster than in reconstituted systems (Dratz et al., 1987; Wagner et al., 1988; Arshavsky et al., 1989, 1991). A number of factors appear to be responsible for Gt GTPase regulation in photoreceptors. One is the activation of Gt GTPase by PDE (Arshavsky et al., 1991; Arshavsky and Bownds, 1992; Pages et al., 1992, 1993), and more specifically it's g subunit acting as a GAP (GTPase activating protein) (Arshavsky and Bownds, 1992; Arshavsky et al. , 1994). Acceleration of Gt GTPase by PDEg requires the presence of ROS membranes. Antonny et al. (1993) have made the point that it is not observed in the soluble complex of Gt -GTP with PDEg. A second factor has been suggested by recent observations that the effect of PDEg becomes more pronounced as ROS membranes are concentrated (Angleson and Wensel, 1994; Arshavsky et al, 1994). The nature of this second mechanism remains unclear.

Two reports suggest that Gt GTPase can be faster than PDE turnoff. Microcalorimetric measurements on ROS have shown GTP and light dependent rapid transients, interpreted as reflecting the time course of Gt activation and hydrolysis (Vuong and Chabre, 1990, 1991). A problem is that the heat responses measured can not be clearly assigned to the hydrolysis of Gt-GTP, rather than dissociation-association reactions of Rh*-Gt or Gt-PDE, as well as some other light-dependent processes. Ting and Ho (1991) report GTPase activity more rapid than PDE inactivation, but do not prove that the burst of GTPase activity observed in their experiments is associated with transducin, nor do they measure GTPase and PDE inactivation under the same conditions. An explicit critique of this work is provided by Antonny et al. (1993).

It remains to be seen whether mechanisms in addition to the transducin GTPase might be involved in PDE inactivation. The rates of Gt GTPase and PDE turnoff measured in the same suspension of bovine ROS coincide (Arshavsky et al., 1989; Angleson and Wensel, 1993). In contrast, Erickson et al. (1992) have recently asserted that PDE inactivation can occur in the absence of GTP hydrolysis when Gt is activated by a non-hydrolyzable analog of GTP. They propose that ROS contain a pool of PDE inhibitor, presumably PDE g -subunits, that inhibit activated PDE prior to the Gt-GTPase reaction. However, their measurements conducted in the presence of GTP, making the Gt-GTPase reaction possible, reveal faster and more complete PDE turnoff when compared with that observed with the non-hydrolyzable analog. The next experiments needed are an examination of the rates of transducin GTPase and PDE turn-off in suspensions of permeabilized, or broken and very concentrated, ROS.

The step of PDE inactivation would seem to be a good locus for light adaptation. A mechanism that increased the rate of PDE turnoff would be expected to lead to a reduction in both photoresponse duration and amplitude during light adaptation. In fact, a regulation of PDE turnoff by cGMP binding to non-catalytic cGMP binding sites on PDE a and b subunits has been demonstrated in recent work (Arshavsky et al., 1991; Arshavsky and Bownds, 1992). When these sites are occupied transducin GTPase and PDE turn-off are slowed. As cGMP concentration falls over its physiological range, from ~ 5 to below 1 mM, the sites empty and GTPase and PDE turn-off are accelerated several fold.

Further work by Arshavsky et al. (1992) has provided evidence that cGMP binding in the PDE non-catalytic sites determines the fashion in which PDE is activated by Gt*. If the non-catalytic sites are occupied by cGMP, Gt activates PDE and remains bound to the PDE heterotetramer. Alternatively, when the sites are empty, Gt* physically removes PDEg from PDEab upon activation. These observations are summarized in the scheme shown below. Stage 1 shows PDE as a heterotetramer in the dark photoreceptor with cGMP bound to its non-catalytic sites. Activation by Gt* removes inhibition of the catalytic sites by PDEg (stage 2). As long as cGMP remains bound to the non-catalytic sites, Gt* remains in a complex with PDE, and the hydrolysis of GTP that causes PDE inactivation is slowed (upper solid arrow). In stage 3, as illumination continues, bound cGMP dissociates and is hydrolyzed. This dissociation might occur from both activated and non-activated PDE. PDE formed with the non-catalytic sites empty (stages 3 and 4) undergoes a different activation-inactivation sequence. Gt* physically removes PDEg, GTPase activity becomes more rapid, and therefore inactivation occurs faster (lower solid arrow). To complete the cycle of Gt, Gta-GDP must dissociate from PDEg and recombine with Gtbg. This process requires the mediation of Gbg (Yamazaki et al., 1990).

How might this process be involved in adaptation? In the dark cGMP levels are high, the non-catalytic cGMP sites on PDE are occupied (see drawing). This slows GTPase activity so that PDE stays on longer, making the light response bigger and longer. When cGMP is lowered by light, cGMP dissociates from non-catalytic sites of activated and possibly non-activated PDE molecules. These PDE molecules that have lost their bound cGMP are inactivated more rapidly upon subsequent activation, making light responses smaller and faster. At first glance, one might suppose that the detachment of PDEg shown in stage 3 would slow inactivation of PDE, because the re-attachment of PDEg to PDEab might be rate limiting. However, direct measurements show that GTP hydrolysis, rather than PDEg re-attachment, is rate limiting under these conditions (Arshavsky, Dumke and Bownds, MS in preparation).

Validation of a model of this sort requires that both the binding constants and off kinetics of the non-catalytic cGMP binding sites be determined. Cote et al. (1994) have recently accomplished this, finding that activation of frog PDE lowers the binding affinity and accelerates the dissociation kinetics of these sites, such that the cGMP drop occurring upon illumination favors release of bound cGMP. The time scale for dissociation, tens of seconds, corresponds to the period during continuous bright illumination of photoreceptors when an increase in the response recovery rate is being observed (Coles and Yamane, 1975; Cervetto et al., 1984; McCarthy and Owen, personal communication).

The mechanism may be limited to lower vertebrates, because bovine PDE does not readily lose bound cGMP in vitro (Gillespie and Beavo, 1989). The non-catalytic cGMP binding sites of vertebrate cone PDEs have binding constants intermediate between the amphibian and vertebrate rod PDEs (Gillespie and Beavo, 1988), and so might potentially use this mechanism to achieve their faster turn-off times. In both cases, it is possible that dissociation occurs more readily for the activated enzyme.

The situation regarding PDEg may become even more complicated if its' recently demonstrated phosphorylation by different protein kinases (Hayashi et al., 1991; Udovichenko et al., 1994; Hayashi, 1994; Tsuboi et al., 1994a,b) is shown to change during the activation-inactivation cycle. Tsuboi et al. (1994a,b) report that a cGMP inhibited kinase can phosphorylate either free PDEg or PDEg in complex with Gta-GTP. The phosphorylated forms lose their ability to bind Gta-GTP, and thus render PDE refractory to activation. While Tsuboi et al. suggest that this PDEg phosphorylation is a mechanism of rapid PDE turnoff independent of GTPase activity of Gta, we think it more likely that it is a means of removing some PDE from the active pool during light adaptation. Experiments are needed to demonstrate that this phosphorylation-dephosphorylation actually occurs during either flash responses or light adaptation.

The drop in cytoplasmic cGMP resulting from PDE activation causes cGMP-gated channels to close. The last step of the phototransduction cascade is the closure of cGMP-gated cationic channels in the ROS plasma membrane. Since their properties are discussed in detail in several recent reviews (Yau and Baylor, 1989; McNaughton, 1990, Kaupp and Koch, 1992; Hsu and Molday, this series) we will provide only a brief account. The amphibian ROS contains approximately 5 x 105 channels, and only about 2% of these are open in the dark-adapted cell. Electrophysiological measurements show that absorption of a single photon leads to a 1-5% reduction in the dark conductance of the plasma membrane, caused by almost complete closure of channels over a length of the ROS of not less than 1mm (Matthews, 1986), or more than 6 mm (Lamb et al., 1981). These numbers are in reasonable correspondence with the calculations above indicating that ~4% of rod cGMP is hydrolyzed during a single photon photoresponse.

Is the channel a site of modification during adaptation? Recent observations of Gordon et al. (1992) suggest that channel sensitivity to cGMP may be regulated by phosphorylation. Plasma membrane patches from frog ROS become more sensitive to cGMP over a period of minutes after their excision, and this sensitivity increase is delayed if protein phosphatase inhibitors are added. Hsu and Molday (1993) have found that the cGMP-gated channel from bovine rods can be modulated by calmodulin. Lowering Ca++ concentration over the range it falls during illumination increases the affinity of the channel for cGMP. Thus less cGMP might be required to re-open the channel during recovery of the photoresponse in the light adapted cell than is required to keep it open in the dark. The physiological relevance of this calmodulin effect is discussed in detail by Hsu and Molday in a companion article in this series. The drop in cytoplasmic Ca++ caused by channel closing activates a negative feedback loop by stimulating guanylate cyclase and cGMP recovery.

Ca++ ions continually enter the dark ROS through cGMP-gated channels and are extruded by a Na+/Ca++,K+ exchanger (see below). The decrease in cytoplasmic calcium that occurs upon channel closing has been shown in numerous studies to be crucial in regulating recovery and adaptation of the photoresponse. Numerous reviews can be consulted on this topic (Pugh and Lamb, 1990; McNaughton, 1990; Stryer, 1991; Kaupp and Koch, 1992; Lagnado and Baylor, 1992; Detwiler and Gray-Keller, 1992; Koch, 1992).

The central electrophysiological observations are that clamping internal Ca++ concentration blocks the gain reduction associated with adaptation (Matthews et al., 1988; Nakatani and Yau, 1988b), and that buffering internal calcium slows the recovery phase of the flash response (Lamb et al., 1986). Biochemical studies on bovine ROS show that guanylate cyclase activity is stimulated by lowering Ca++ from 200 nM to 50 nM (Koch and Stryer, 1988). This Ca++ sensitivity was first thought to be mediated by the 26kDa Ca++ binding protein, recoverin (Dizhoor et al., 1991; Lambrecht and Koch, 1991). Recoverin, however has since been shown to be a regulator of rhodopsin phosphorylation rather than quanylate cyclase (Hurley et al, 1993; see above). Two groups have now isolated small protein(s) that confer Ca++ sensitivity on cyclase (Gorczyca et al., 1994; Baehr et al., 1994; Dizhoor et al., 1994).

These observations are all consistent with the idea that activation of cyclase caused by the lowering of calcium can accelerate cGMP resynthesis and the restoration of dark conductance. Several quantitative models of the photoresponse have included the calcium feedback loop and reproduce electrophysiological recordings very closely (Sneyd and Tranchina, 1989; Forti et al., 1989; Tranchina et al., 1991; Tamura et al., 1991). These models, however, do not exclude the possibility of further regulatory loops, and in fact Koutalos, Nakatani and Yau (personal communication) have established the importance of both the cyclase feedback loop and the PDE attenuation that occurs as a fall in Ca++ levels accelerates rhodopsin phosphorylation and inactivation mentioned above. They used the truncated rod outer segment preparation to measure the Ca++-dependence of the cGMP-gated channel, the quanylate cyclase, and the steady-state PDE activity elicited by a step of light. They were able to account for the steady-state response and sensitivity of the intact rod as a function of background light intensity. In terms of relative contribution to adaptation to background light, they found that the Ca++ modulation of the cyclase is most important at low background light intensities and modulation of light-activated PDE contributes at higher light levels. The Ca++ modulation of the channel was found to add relatively little to light adaptation.

Several groups have used fluorescent probes (Ratto et al., 1988; Younger et al., 1992; McCarthy et al., 1993; Gray-Keller and Detwiler, 1994) or aequorin (Lagnado et al.,1992) to measure Ca++ concentration changes that occur on illumination. The most recent estimates judge dark Ca++ concentration to be approximately 500 nM, falling to ~50 nM upon saturating illumination. The dark level corresponds to ~80 Ca++ ions per disk face, falling below ~8 upon illumination. This represents a very small fraction of the capacity of the Na+/Ca++,K+ exchanger (McNaughton, 1990). The fall occurs in two phases, the first in less than one second and the next with a half-time of ~5 seconds. Changes in cytosolic free Ca++, sensitivity and fractional current suppression follow a broadly similar form during steady state exposure to different background light intensities.

It is unlikely that calcium is the sole regulator of adaptation processes. Photoresponses with normal kinetics can be measured in Ca++ depleted ROS, incubated in 10 nM Ca++ and calcium ionophore (Nicol et al., 1987), and there would appear to be mechanisms capable of reducing the gain of the photoresponse and speeding its recovery at higher background intensities in the absence of significant changes in cytoplasmic calcium (Nicol and Bownds, 1989). The cGMP feedback mentioned above would be one candidate for such a mechanism. Rispoli and Detwiler (1991) have demonstrated that increasing internal Ca++ concentration speeds, rather than slows, response recovery. In an earlier report (1989) they also noted that the recovery kinetics of the light response were independent of the dark current level that presumably regulates internal Ca++.

Can the most important sites of light adaptation be specified? In summary, recent experiments have provided evidence for pathways by which the primary transduction messenger, cGMP, can exert feedback control over both its synthesis and light-induced degradation. The light induced decrease in cGMP can damp further decreases by leading to stimulation of guanylate cyclase and accelerating the inactivation of light-activated PDE: A central question is whether one of the mechanisms we have mentioned plays a predominant role in adaptation, or whether equally important multiple attenuations occur at each step of the transduction cascade. The fact that all of the turn-off reactions of the cascade have now been implicated as sites of Ca++ or cGMP regulation suggests that the latter possibility is more likely. The data suggest that light adaptation consists of enhancement of all of the turnoff and restoration processes accompanying the photoresponse, starting with rhodopsin phosphorylation and ending with the putative modulation of the plasma membrane channel by Ca++ or phosphorylation. We think it likely that a sequence of reactions are recruited as background light intensities increase: 1). Ca++ feedback on guanylate cyclase; 2). Ca++ feedback on rhodopsin phosphorylation; 3). cGMP feedback on the lifetime of PDE; 4.) phosphorylation of unbleached rhodopin.

Three reactions appear to be most central for turnoff of the photoresponse: rhodopsin quenching, PDE inactivation and guanylate cyclase activation. The data presently available do not permit us to argue strongly for one of these being most central. The crucial balance between the enzymes that set cGMP concentration during recovery of the photoresponse might be considered in several ways:

Several reviews emphasize cyclase activation rather than PDE inactivation as central in cGMP and conductance restoration after a flash of light. We think it most likely that the function of cyclase acceleration is to avoid a delay between PDE turnoff and cGMP concentration restoration. Such a delay is most likely the cause of the lengthening of the flash response that occurs when calcium buffers are injected into the ROS cytoplasm. Is cyclase activation sufficient to contribute to adaptation during background light? The >300-fold activation of PDE that occurs upon illumination (Arshavsky et al. 1992, and see above) needs to be compensated by cyclase activation. Although the maximum cyclase activation measured in vitro has been < 10-fold (reviewed in Stryer, 1991) this activation (mediated by diffusible Ca++) could be expected to spread much further than the disks containing activated PDE molecules. This suggests that at lower levels of illumination, where cyclase is activated over a significantly larger cell volume than PDE, cyclase activation may be sufficient to be part of the mechanism that counters PDE activation during light adaptation. This suggestion is compatible with the finding of Koutalos, Nakatani and Yau, mentioned above, that modulation of cyclase activity by Ca++ is most important at lower background light intensities.

Another consideration is to determine to what extent Rh* turnoff versus PDE inactivation is rate limiting for terminating cyclic GMP hydrolysis. PDE inactivation would be rate limiting if Rh* turnoff were significantly faster than the rate of Gt-GTPase that causes termination of PDE activation. Conversely, Rh* decay would become limiting if its rate were slower than that of the GTPase. Current biochemical data suggest that these processes are occurring at similar times, so that each might be modulated during light adaptation. Further progress in understanding which of these reactions is most important requires the development of techniques that measure them on the time scale of the photoresponse at dim light intensities. It would not be surprising to find controls that are redundant and increase the reliability of the system. Studies on this system are repeating the history of the discoveries of basic catabolic and anabolic pathways in the 1950's and 1960's. After establishing a basic pathway, even more effort is being directed towards describing the relevant array of feedback and other regulatory controls.

Acknowledgment: This manuscript was prepared with support from N.I.H. grants EY-00463 and EY-10336. We are grateful for comments on the manuscript of Drs. Mark Gray-Keller, Peter Detwiler, Marc Chabre and several anonymous reviewers. The article was completed in October 1993 and some more recent references were added in a minor revision in July 1994.

REFERENCES

Anant, J.S., Ong, O.C., Xie, H., Clarke, S., O'Brien, P.J. & Fung, B.K.-K. (1992) In vivo differential prenylation of retinal cyclic GMP phosphodiesterase catalytic subunits. The Journal of Biological Chemistry 267:687-690.

Anant, J.S. & Fung, B.K.-K. (1992) In vivo farnesylation of rat rhodopsin kinase. Biochemical and Biophysical Research Communications 183:468-473. Angleson, J.K. & Wensel, T.G. (1993) A GTPase-accelerating factor for transducin, distinct from its effector cGMP phosphodiesterase, in rod outer segment membranes. Neuron 11:939-949.

GTPase accelerating protein activity by the inhibitory subunit of cGMP phosphodiesterase. The Journal of Biological Chemistry 269:16290-16296.

Antonny, B., Otto-Bruc, A., Chabre, M. & Minh Vuong, T. (1993) GTP hydrolysis by purified -subunit of transducin and its complex with the cyclic GMP phosphodiesterase inhibitor. Biochemistry 32:8646-8653.

Arshavsky, V.Y., Dizhoor, A.M., Shestakova, I.K. & Philippov, P.P. (1985) The effect of rhodopsin phosphorylation on the light-dependent activation of phosphodiesterase from bovine rod outer segments. FEBS Letters 181:264-266.

Arshavsky, V.Y., Antoch, M.P., Dizhoor, A.M., Rakhilin, S.V. & Philippov, P.P. (1987) Interaction of visual transduction enzymes in a detergent solution. In: Retinal Proteins, ed. Ovchinnikov, Y.A. VNU Science Press. Utrecht, The Netherlands, 497-503.

Arshavsky, V.Y., Antoch, M.P., Lukjanov, K.A. & Philippov, P.P. (1989) Transducin GTPase provides for rapid quenching of the cGMP cascade in rod outer segments. FEBS Letters 250:353-356.

Arshavsky, V.Y., Gray-Keller, M.P. & Bownds, M.D. (1991) cGMP suppresses GTPase activity of a portion of transducin equimolar to phosphodiesterase in frog rod outer segments. Light-induced cGMP decreases as a putative feedback mechanism of the photoresponse. The Journal of Biological Chemistry 266:18530-18537.

Arshavsky, V.Y., Dumke, C.L. & Bownds, M.D. (1992) Non-catalytic cGMP binding sites of amphibian rod cGMP phosphodiesterase control interaction with its' inhibitory gamma-subunits - a putative regulatory mechanisms of the rod photoresponse. The Journal of Biological Chemistry 267:24501-24507.

Arshavsky, V.Y., Dumke, C.L., Zhu, Y., Artemyev, N.O., Skiba, N.P., Hamm, H.E. & Bownds, M.D. (1994) Regulation of transducin GTPase activity in bovine rod outer segments. The Journal of Biological Chemistry 269:in press.

Arshavsky, V.Y. & Bownds, M.D. (1992) Regulation of deactivation of photoreceptor G protein by its target enzyme and cGMP. Nature 357:416-417.

Baehr, W., Morita, E.A., Swanson, R.J. & Applebury, M.L. (1982) Characterization of Bovine Rod Outer Segment G-protein. The Journal of Biological Chemistry 257:6452-6460.

Baehr, W., Gorczyca, W., Subbaraya, I., Ohguro, H., Johnson, R.S., Walsh, K.A., Gray-Keller, M.P., Detwiler, P.B., Crabb, J. & Palczewski, K. (1994) Isolation, cloning, and functional characterization of bovine retina guanylate cyclase activating protein (GCAP). Investigative Ophthalmology and Visual Science 35:1486.

Baylor, D.A., Lamb, T.D. & Yau, K.-W. (1979a) The membrane current of single rod outer semgments. The Journal of Physiology (London) 288:589-611.

Baylor, D.A., Lamb, T.D. & Yau, K.-W. (1979b) Responses of retinal rods to single photons. The Journal of Physiology (London) 288:613-634.

Baylor, D.A., Matthews, G. & Yau, K.-W. (1980) Two components of electrical dark noise in toad retinal rod outer segments. The Journal of Physiology (London) 309:591-621.

Baylor, D.A. & Hodgkin, A.L. (1974) Changes in time scale and sensitivity in turtle photoreceptors. The Journal of Physiology (London) 242:729-758.

Bennett, N. & Clerc, A. (1989) Activation of cGMP phosphodiesterase in retinal rods: Mechanism of interaction with the GTP-binding protein (transducin). Biochemistry 28:7418-7424.

Bennett, N. & Sitaramayya, A. (1988) Inactivation of photoexcited rhodopsin in retinal rods; the roles of rhodopsin kinase and 48-KDa protein (arrestin). Biochemistry 27:1710-1715.

Biernbaum, M.S., Binder, B.M. & Bownds, M.D. (1991) Dim background light and Cerenkov radiation from 32P block reversal of rhodopsin phosphorylation in intact frog retinal rods. Visual Neuroscience 7:499-503.

Biernbaum, M.S. & Bownds, M.D. (1985) Light-induced changes in GTP and ATP in Frog rod photoreceptors. Comparison with recovery of dark current and light sensitivity during dark adaptation. The Journal of General Physiology 85:107-121.

Binder, B.M., Brewer, E. & Bownds, M.D. (1989) Stimulation of protein phosphorylations in frog rod outer segments by protein kinase activators. Suppression of light-induced changes in membrane current and cGMP by protein kinase C activators. The Journal of Biological Chemistry 264:8857-8864.

Binder, B.M., Biernbaum, M.S. & Bownds, M.D. (1990) Light activation of one rhodopsin molecule causes the phosphorylation of hundreds of others. A reaction observed in electropermeabilized frog rod outer segments exposed to dim illumination. The Journal of Biological Chemistry 265:15333-15340.

Bourne, H.R., Sanders, D.A. & McCormick, F. (1990) The GTPase superfamily: a conserved switch for diverse cell functions. Nature 348:125-132.

Bourne, H.R., Sanders, D.A. & McCormick, F. (1991) The GTPase superfamily:conserved structure and molecular mechanism. Nature 349:117-127.

Bownds, D., Dawes, J., Miller, J. & Stahlman, M. (1972) Phosphorylation of frog photoreceptor membrnaes induced by light. Nature New Biology 237:125-127.

Bownds, M.D., Arshavsky, V.Y., Calvert, P.D. & Dumke, C.I. (1992) Rod outer segment concentration and structure influence the apparent kinetics of cGMP phosphodiesterase. Investigative Ophthalmology and Visual Science 33:873.

Bownds, M.D. & Thomson, D. (1988) Control of the sensitivity and kinetics of the vertebrate photoresponse during light adaptation - calculations from simple molecular models. In: Molecular Physiology of Retinal Proteins, ed. Hara, T. Japan Scientific Soc. Press. Osaka, 241-246.

Breer, H. & Boekhoff, I. (1992) Second messenger signalling in olfaction. Current Opinion in Neurobiology 2:439-443.

Buczylko, J., Gutmann, C. & Palczewski, K. (1991) Regulation of rhodopsin kinase by autophosphorylation. Proceedings of the National Academy of Sciences USA 88:2568-2572.

Cervetto, L., Torre, V., Pasino, E., Marroni, P. & Capovilla, M. (1984) Recovery from light-desensitization in toad rods. In: Photoreceptors, ed. Borsellino, A. & Cervetto, L. Plenum Press. New York and London, 159-175.

Chabre, M. & Deterre, P. (1989) Molecular mechanism of visual transduction. European Journal of Biochemistry 179:255-266.

Charbonneau, H., Prusti, R.K., LeTrong, , Sonnenburg, W.K., Mullaney, P.J., Walsh, K.A. & Beavo, J.A. (1990) Identification of a noncatalytic cGMP-binding domain conserved in both the cGMP-stimulated andd photoreceptor cyclic nucleotide phosphodiesterase. Proceedings of the National Academy of Sciences USA 87:288-292.

Chen, C.-K. & Hurley, J.B. (1994) Calcium dependent recoverin/rhodopsin kinase interaction. Investigative Ophthalmology and Visual Science 35:1485.

Clack, J.W. & Pepperberg, D.R. (1982) Desensitization of skate photoreceptor by bleaching and background light. The Journal of General Physiology 80:863-883.

Coles, J.A. & Yamane, S. (1975) Effects of adapting lights on the time course of the receptor potential of the anuran retinal rod. The Journal of Physiology (London) 247:189-207.

Cornwall, M.C. & Fain, G.L. (1992) Bleaching of rhodopsin in isolated rods causes a sustained activation of PDE and Cyclase which is reversed by pigment regeneration. Investigative Ophthalmology and Visual Science 33:1103.

Cote, R.H., Bownds, M.D. & Arshavsky, V.Y. (1994) cGMP binding sites on photoreceptor phosphodiesterase: Role in feedback regulation of visual transduction. Proceedings of the National Academy of Sciences USA 91:4845-4849.

Cote, R.H. & Brunnock, M.A. (1993) Intracellular cGMP concentration in rod photoreceptors is regulated by binding to high and moderate affinity cGMP binding sites. The Journal of Biological Chemistry 268:17190-17198.

Dawis, S.M., Graeff, R.M., Heyman, R.A., Walseth, T.F. & Goldberg, N.D. (1988) Regulation of cyclic GMP metabolism in toad photoreceptors. The Journal of Biological Chemistry 263:8771-8785.

Dawis, S.M. (1991) A molecular basis for Weber's law. Visual Neuroscience 7:285-320.

Deterre, P., Bigay, J., Forguet, F., Robert, M. & Chabre, M. (1988) cGMP of phosphodiesterase of retinal rods is regulated by two inhibitory subunits. Proceedings of the National Academy of Sciences USA 85:2424-2428.

Detwiler, P.B. & Gray-Keller, M.P. (1992) Some unresolved issues in the physiology and biochemistry of phototransduction. Current Opinion in Neurobiology 2:433-438.

Dizhoor, A.M., Ray, S., Kumar, S., Niemi, G., Spencer, M., Brolley, D., Walsh, K.A., Philipov, P.P., Hurley, J.B. & Stryer, L. (1991) Recoverin: A calcium sensitive activator of retinal rod guanylate cyclase. Science 251:915-918.

Dizhoor, A.M., Lowe, D.G., Olshevskaya, E.V., Laura, R.P. & Hurley, J.B. (1994) The human photoreceptor membrane guanylyl cyclase, RetGC, is present in outer segments and is regulated by calcium and a soluble activator. Neuron 12:1345-1352.

Dratz, E.A., Lewis, J.W., Schaechter, L.E., Parker, K.R. & Kliger, D.S. (1987) Retinal Rod GTPase turnover rate increases with concentration: a key to the control of visual excitation? Biochemical and Biophysical Research Communications 146:379-386.

Dumke, C.L., Arshavsky, V.Y., Calvert, P.D., Pugh, E.N. & Bownds, M.D. (1994) Rod outer segment structure influences the apparent kinetic parameters of cGMP phosphodiesterase. The Journal of General Physiology 103:1071-1098.

visual transduction without guanosine triphosphate hydrolysis by G protein. Science 257:1255-1258.

Fain, G.L. (1976) Sensitivity of toad rods; dependence on wavelength and background illumination. The Journal of Physiology (London) 261:71-101.

Fain, G.L. & Lisman, J.E. (1993) Photoreceptor degeneration in vitamin A deprivation and retinitis pigmentosa: The equivalent light hypothesis. Experimental Eye Research 57:335-340.

Firestein, S. (1991) A noseful of odor receptors. Trends in Neuroscience 14:270-272.

Flaherty, K.M., Zozulya, S., Stryer, L. & McKay, D.B. (1993) Three-dimensional structure of recoverin, a calcium sensor in vision. Cell 75:709-716.

Fliesler, S.J. & Anderson, R.E. (1983) Chemistry and metabolism of lipids in the vertebrate retina. Progress in Lipid Research 22:79-131.

Forti, S., Menini, A., Rispoli, G. & Torre, V. (1989) Kinetics of phototransduction in retinal rods of the newt Triturus cristatus. The Journal of Physiology (London) 419:265-295.

Fowles, C., Charma, R. & Akhtar, M. (1988) Mechanistic studies on the phosphorylation of photoexcited rhodopsin. FEBS Letters 238:56-60.

Fukada, Y., Takao, T., Ohguor, H., Yoshizawa, T., Akino, T. & Shimonishi, Y. (1990) Farnesylated gamma subunit of photoreceptor G protein indispensable for GTP-binding. Nature 346:658-660.

Fung, B.B.-K., Hurley, J.B. & Stryer, L. (1981) Flow of information in the light-triggered cyclic nucleotide cascade of vision. Proceedings of the National Academy of Sciences USA 78:152-156.

Fung, B.K.-K., Young, J.H., Yamane, H.K. & Griswold-Prenner, I. (1990) Subunit stoichiometry of retinal rod cGMP phosphodiesterase. Biochemistry 29:2657-2664.

Gillespie, P.G. & Beavo, J.A. (1988) Characterization of a bovine cone photoreceptor phosphodiesterase purified by cyclic GMP-sepharose chromatography. The Journal of Biological Chemistry 263:8133-8141.

Gillespie, P.G. & Beavo, J.A. (1989) cGMP is tightly bound to bovine retinal rod phosphodiesterase. Proceedings of the National Academy of Sciences USA 86:4311-4315.

Gorczyca, W.A., Gray-Keller, M.P., Detwiler, P.B. & Palczewski, K. (1994) Purification and physiological evaluation of a guanylate cyclase activating protein from retinal rods. Proceedings of the National Academy of Sciences USA 91:4014-4018.

Gordon, S.E., Brautigan, D.L. & Zimmerman, A.L. (1992) Protein phosphatases modulate the apparent agonist affinity of the light-regulated ion channel in retinal rods. Neuron 9:739-748.

Gray-Keller, M.P., Biernbaum, M.S. & Bownds, M.D. (1990) Transducin activation in electropermeabilized frog rod outer segments is highly amplified, and a portion equivalent to phosphodiesterase remains membrane-bound. The Journal of Biological Chemistry 265:15323-15332.

Gray-Keller, M.P., Polans, A.S., Palczewski, K. & Detwiler, P.B. (1993) The effect of recoverin-like calcium-binding proteins on the photoresponse of retinal rods. Neuron 10:523-531.

Gray-Keller, M.P. & Detwiler, P.B. (1994) Intracellular calcium measurements in isolated rod photoreceptors. Investigative Ophthalmology and Visual Science 35:1486.

Groskoph, S.L., Hammett, D.J. & Cote, R.H. (1992) Nucleotide energy metabolism in electropermeabilized rod photoreceptors. Investigative Ophthalmology and Visual Science 33:1008.

Hamm, H. (1990) Regulation by light of cyclic nucleotide-dependent protein kinases and their substrates in frog rod outer segments. The Journal of General Physiology 95:545-567.

Hamm, H.E. & Bownds, M.D. (1986) Protein complement of rod outer semgments of frog retina. Biochemistry 25:4512-4523.

Hargrave, P.A. & McDowell, J.H. (1992) Rhodopsin and phototransduction: A model system for G protein-linked receptors. FASEB Journal 6:2323-2331.

Hayashi, F., Lin, G.Y., Matsumoto, H. & Yamazaki, A. (1991) Phosphatidylinositol-stimulated phosphorylation of an inhibitory subunit of cGMP phosphodiesterase in vertebrate rod photoreceptors. Proceedings of the National Academy of Sciences USA 88:4333-4337.

Hayashi, F. (1994) Light-dependent in vivo phosphorylation of an inhibitory subunit of cGMP-phosphodiesterase in frog rod photoreceptor outer segments. FEBS Letters 338:203-206.

Heck, M. & Hofmann, K.P. (1993) G-protein-effector coupling: A real-time light-scattering assay for transducin-phosphodiesterase interaction. Biochemistry 32:8220-8227.

Hermolin, J., Karell, M.A., Hamm, H.E. & Bownds, M.D. (1982) Calcium and cyclic GMP regulation of light-sensitive protein phosphorylation in frog photoreceptor membranes. The Journal of General Physiology 79:633-655.

Hodgkin, A.L. & Nunn, B.J. (1988) Control of light-sensitive current in salamander rods. The Journal of Physiology (London) 403:439-471.

Hofmann, K.P., Pulvermvller, A., Buczylko, J., Van Hooser, P. & Palczewski, K. (1992) The role of arrestin and retinoids in the regeneration pathway of rhodopsin. The Journal of Biological Chemistry 267:15701-15706.

Hsu, Y.-T. & Molday, R.S. (1993) Modulation of the cGMP-gated channel of rod photoreceptor cells by calmodulin. Nature 361:76-79.

Hurley, J.B. (1992) Signal Transduction Enzymes of Vertebrate Photoreceptors. Journal of Bioenergetics and Biomembranes 24:219-226.

Hurley, J.B., Dizhoor, A.M., Ray, S. & Stryer, L. (1993) Recoverin's role: Conclusion withdrawn. Science 260:740.

Hurley, J.B. & Stryer, L. (1982) Purification and characterization of the gamma regulatory subunit of the cGMP phosphodiesterase from retinal rod outer segments. The Journal of Biological Chemistry 257:11094-11099.

Jin, J., Crouch, R.K., Corson, D.W., Katz, B.M., MacNichol, E.F & Cornwall, M.C. (1993) Noncovalent Occupancy of the Retinal-Binding pocket of opsin diminishes bleaching adaptation of retinal cones. Neuron 11:513-522.

Kahlert, M., Pepperberg, D.R. & Hofmann, K.P. (1990) Effect of bleached rhodopsin on signal amplification in rod visual receptors. Nature 345:537-539.

Kaupp, U.B. & Koch, K.-W. (1992) Role of cGMP and Ca2+ in vertebrate photoreceptor excitation and adaptation. Annual Review of Physiology 54:153-175.

Kawamura, S., Takamatsu, K. & Kitamura, K. (1992) Purification and characterization of S-modulin, a calcium-dependent regulator on cGMP phosphodiesterase in frog rod photoreceptors. Biochemical and Biophysical Research Communications 186:411-417.

Kawamura, S. (1993) Rhodopsin phosphorylation as a mechanism of cyclic GMP phosphodiesterase regulation by S-modulin. Nature 362:855-857.

Kawamura, S. & Bownds, M.D. (1981) Light Adaptation of the Cyclic GMP Phosphodiesterase of Frog Photoreceptor Membranes Mediated by ATP and Calcium Ions. The Journal of General Physiology 77:571-591. cyclic GMP phosphodiesterase by a protein from frog retinal rods. Nature 349:420-423.

Khorana, H.G. (1992) Rhodopsin, photoreceptor of the rod cell. The Journal of Biological Chemistry 267:1-4.

Koch, K.-W. (1992) Biochemical mechanism of light adaptation in vertebrate photoreceptors. Trends in Biochemical Science 17:307-311.

Koch, K.-W. & Stryer, L. (1988) Highly cooperative feedback control of retinal rod guanylate cyclase by calcium ions. Nature 334:64-66.

Korenbrot, J.I., Brown, D.T. & Cone, R.A. (1973) Membrane characteristics and osmotic behavior of isolated rod outer segments. The Journal of Cell Biology 56:389-398.

Koutalos, Y. & Yau, K.-W. (1993) A rich complexity emerges in phototransduction. Current Opinion in Neurobiology 3:513-519.

Lagnado, L., Cervetto, L. & McNaughton, P.A. (1992) Calcium homeostasis in the outer segments of retinal rods from the tiger salamander. The Journal of Physiology (London) 455:111-142.

Lagnado, L. & Baylor, D. (1992) Signal flow in visual transduction. Neuron 8:995-1002.

Lagnado, L. & Baylor, D.A. (1994) Calcium controls light-triggered formation of catalytically active rhodopsin. Nature 367:273-277.

Lamb, T.D., McNaughton, P.A. & Yau, K.-W. (1981) Spatial spread of activation and background desensitization in toad rod outer segments. The Journal of Physiology (London) 319:463-496.

Lamb, T.D. (1984) Effects of temperature changes on toad rod photocurrents. The Journal of Physiology (London) 346:557-578.

Lamb, T.D., Matthews, H.R. & Torre, V. (1986) Incorporation of calcium buffers into salamander retinal rods; a rejection of the calcium hypothesis of phototransduction. The Journal of Physiology (London) 372:315-349.

Lamb, T.D. & Pugh, E.N.,Jr. (1992) A quantitative account of the activation steps involved in phototransduction in amphibian photoreceptors. The Journal of Physiology (London) 449:719-758.

Lambrecht, H.-G. & Koch, K.-W. (1991) A 26 kd calcium binding protein from bovine rod outer segments as modulator of photoreceptor guanylate cyclase. EMBO Journal 10:793-798.

Lambright, D.G., Noel, J.P., Hamm, H.E. & Sigler, P.B. (1994) Structural determinants for activation of the alpha subunit of a heterotrimeric G protein. Nature 369:621-628.

Lee, R.H., Lieberman, B.S. & Lolley, R.N. (1987) A novel complex from bovine visual cells of a 33,000-dalton phosphoprotein with beta and gamma transducin: purification and subunit structure. Biochemistry 28:3983-3990.

Lee, R.H., Whelan, J.P., Lolley, R.N. & McGinnis, J.F. (1988) The photoreceptor-specific 33 kDa phosphoprotein of mammalian retina: generation of monospecific antibodies and localization by immunocytochemistry. Experimental Eye Research 46:829-840.

Lee, R.H., Fowler, A., McGinnis, J.F., Lolley, R.N. & Craft, C.M. (1990a) Amino acid and cDNA sequence of bovine phosducin, a soluble phosphoprotein from photoreceptor cells. The Journal of Biological Chemistry 265:15867-15873.

Lee, R.H., Lieberman, B.S. & Lolley, R.N. (1990b) Retinal accumulation of the phosducin/Tgamma and transducin complexes in developing normal mice and in mice and dogs with inherited retinal degeneration. Experimental Eye Research 51:325-333.

Li, H., Volpp, K. & Applebury, M.L. (1990) Bovine cone photoreceptor cGMP phosphodiesterase structure deduced from a cDNA clone. Proceedings of the National Academy of Sciences USA 87:293-297.

Liebman, P.A., Parker, K.R. & Dratz, E.A. (1987) 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.

Liebman, P.A. & Entine, G. (1968) Visual pigments of frog and tadpole. Vision Research 8:761-775.

Liebman, P.A. & Pugh, E.N. (1979) The control of phosphodiesterase in rod disk membranes: kinetics, possible mechanisms and significance for vision. Vision Research 19:375-380.

Liebman, P.A. & Pugh, E.N. (1980) ATP mediates rapid reversal of cyclic GMP phosphodiesterase activation in visual receptor membranes. Nature 287:734-737.

Matthews, G. (1986) Spread of the light response along the rod outer segment: an estimate from patch-clamp recordings. Vision Research 26:535-541.

Matthews, H.R., Murphy, R.L.W., Fain, G.L. & Lamb, T.D. (1988) Photoreceptor light adaptation is mediated by cytoplasmic calcium concentration. Nature 334:67-69.

McCarthy, S.T., Younger, J.P. & Owen, W.G. (1993) Disc structure, buffering and kinetics of cytosolic free calcium in bullfrog rod outer segments. Investigative Ophthalmology and Visual Science 34:1327.

McDowell, J.H., Nawrocki, J.P. & Hargrave, P.A. (1993) Phosphorylation sites in bovine rhodopsin. Biochemistry 32:4968-4974.

McNaughton, P.A. (1990) Light response of vertebrate photoreceptors. Physiological Reviews 70:847-884.

Miller, J.L., Fox, D.A. & Litman, B.J. (1986) Amplification of phosphodiesterase activation is greatly reduced by rhodopsin phosphorylation. The Journal of Biological Chemistry 25:4983-4988.

Nakatani, K., Tamura, T. & Yau, K.-W. (1991) Light adaptation in retinal rods of the rabbit and two other nonprimate mammals. The Journal of General Physiology 97:413-435.

Nakatani, K. & Yau, K.-W. (1988a) Guanosine 3'-5'-cyclic monophosphate-activated conductance studied in a truncated rod outer segment of the toad. The Journal of Physiology (London) 395:731-753.

Nakatani, K. & Yau, K.-W. (1988b) Calcium and light adaptation in retinal rods and cones. Nature 334:69-71.

Nathans, J. (1992) Rhodopsin: Structure, function, and genetics. Biochemistry 31:4923-4931.

Neubert, T.A., Johnson,R.S., Hurley,J.B. & Walsh,K.A. (1992) The rod transducin alpha subunit amino terminus is heterogeneously fatty acylated. The Journal of Biological Chemistry 267:18274-18277.

Newton, A.C. & Williams, D.S. (1991) Involvement of protein kinase C in the phosphorylation of rhodopsin. The Journal of Biological Chemistry 266:17725-17728.

Newton, A.C. & Williams, D.S. (1993) Rhodopsin is the major in situ substrate of protein kinase C in rod outer segments of photoreceptors. The Journal of Biological Chemistry 268:18181-18186.

Nicol, G.D., Kaupp, U.B. & Bownds, M.D. (1987) Transduction persists in rod photoreceptors after depletion of intracellular calcium. The Journal of General Physiology 89:297-319.

Nicol, G.D. & Bownds, M.D. (1989) Calcium regulates some, but not all, aspects of light adaptation in rod photoreceptors. The Journal of General Physiology 94:233-259.

Noel, J.P., Hamm, H.E. & Sigler, P.B. (1993) The 2.2  crystal structure of transducin- complexed with GTPgammaS. Nature 366:654-663.

Ohguro, H., Fukada, Y., Takao, T., Shimonishi, Y., Yoshizawa, T. & Akino, T. (1991) Carboxyl methylation and farnesylation of transducin gamma-subunit synergistically enhance its coupling with metarhodopsin II. EMBO Journal 10:3669-3674.

Ohguro, H., Palczewski, K., Ericsson, L.H., Walsh, K.A. & Johnson, R.S. (1993) Sequential phosphorylation of rhodopsin at multiple sites. Biochemistry 32:5718-5724.

Pag\s, F., Deterre, P. & Pfister, C. (1992) Enhanced GTPase activity of transducin when bound to cGMP phosphodiesterase in bovine retinal rods. The Journal of Biological Chemistry 267:22018-22021.

Pag\s, F., Deterre, P. & Pfister, C. (1993) Enhancement by phosphodiesterase subunits of the rate of GTP hydrolysis by transducin in bovine retinal rods. Essential role of the phosphodiesterase catalytic core. The Journal of Biological Chemistry 268:26358-26364.

Palczewski, K., McDowell, J.H. & Hargrave, P.A. (1988a) Purification and characterization of rhodopsin kinase. The Journal of Biological Chemistry 263:14067-14073. Palczewski, K., McDowell, J.H. & Hargrave, P.A. (1988b) Rhodopsin kinase: substrate specificity and factors that influence activity. Biochemistry 27:2306-2312.

Palczewski, K., Buczylko, J., Kaplan, M.W., Polans, A.S. & Crabb, J.W. (1991) Mechanism of rhodopsin kinase activation. The Journal of Biological Chemistry 266:12949-12955.

Palczewski, K., Rispoli, G. & Detwiler, P.B. (1992) The influence of arrestin (48K protein) and rhodopsin kinase on visual transduction. Neuron 8:117-126.

Papac, D.I., Oatis, J.E.,Jr., Crouch, R.K. & Knapp, D.R. (1993) Mass spectrometric identification of phosphorylation sites in bleached bovine rhodopsin. Biochemistry 32:5930-5934.

Pepperberg, D.R., Cornwall, M.C., Kahlert, M., Hofmann, K.P., Jin, J., Jones, G.J. & Ripps, H. (1992) Light-dependent delay in the falling phase of the retinal rod photoresponse. Visual Neuroscience 8:9-18.

Pepperberg, D.R., Jin, J. & Jones, G.J. (1994) Modulation of transduction gain in light adaptation of retinal rods. Visual Neuroscience 11:53-62.

P]rez-Sala, D., Tan, E.W., Cagada, F.J. & Rando, R.R. (1991) Methylation and demethylation reactions of guanine nucleotide-binding proteins of retinal rod outer segments. Proceedings of the National Academy of Sciences USA 88:3043-3046.

Pfister, C., Bennett, N., Bruckert, F., Catty, P., Clerc, A., Pag\s, F. & Deterre, P. (1993) Interactions of a G-protein with its effector: Transducin and cGMP phosphodiesterase in retinal rods. Cellular Signalling 5:235-251.

Polans,A.S., Hermolin,J. & Bownds,M.D. (1979) Light-induced phosphorylation of two proteins in frog rod outer segments. The Journal of General Physiology 74:595-613.

Pugh, E.N.,Jr. & Lamb, T.D. (1990) Cyclic GMP and calcium: The internal messengers of excitation and adaptation in vertebrate photoreceptors. Vision Research 30:1923-1948.

Pugh, E.N.,Jr. & Lamb, T.D. (1993) Amplification and kinetics of the activation steps in phototransduction. Biochimica et biophysica Acta Bio-Energetics 1141:111-149.

Pulvermuller, A., Palczewski, K. & Hofmann, K.P. (1993) Interaction between photoactivated rhodopsin and its kinase: Stability and kinetics of complex formation. Biochemistry 32:14082-14088.

Ratto, G.M., Payne, R., Owen, W.G. & Tsien, R.Y. (1988) The concentration of cytosolic free calcium in vertebrate rod outer segments measured with Fura-2. Journal of Neuroscience 8:3240-3246.

Rispoli, G., Sather, W.A. & Detwiler, P.B. (1993) Visual transduction in dialysed detached rod outer segments from lizard retina. The Journal of Physiology (London) 465:513-537.

Rispoli, G. & Detwiler, P.B. (1989) Light adaptation in Gecko rods may involve changes in both the initial and terminal sage of the transduction cascade. Biophysical Journal 55:380.

Rispoli, G. & Detwiler, P.B. (1991) Interactions Between Calcium and cGMP in Dialyzed Detached Retinal Rod Outer Segments. In: Signal Transduction in Photoreceptor Cells, ed. Hargrave, P.A. & Hofmann, K.P. Springer Verlag. Berlin

Robinson, P.R., Kawamura, S., Abramson, B. & Bownds, M.D. (1980) Control of the Cyclic GMP Phosphodiesterase of Frog Photoreceptor Membranes. The Journal of General Physiology 76:631-645.

Robinson, P.R., Cohen, G.B., Zhukovsky, E.A. & Oprian, D.D. (1992) Constitutively active mutants of rhodopsin. Neuron 9:719-725.

Ronnett, G.V. & Snyder, S.H. (1992) Molecular messengers of olfaction. Trends in Neuroscience 15:508-513.

Roth, N.S., Campbell, P.T., Caron, M.G., Lefkowitz, R.J. & Lohse, M.J. (1991) Comparative rates of desensitization of -adrenergic receptors by the -adrenergic receptor kinase and the cyclic AMP-dependent protein kinase. Proceedings of the National Academy of Sciences USA 88:6201-6204.

Sather, W.A. & Detwiler, P.B. (1987) Intracellular biochemical manipulation of phototransduction in detached rod outer segments. Proceedings of the National Academy of Sciences USA 84:9290-9294.

Shapley, R. & Enroth-Cugell, C. (1984) Visual adaptation and retinal gain controls. Progress in Retinal Research 3:263-346.

Shuster, T.A. & Farber, D.B. (1984) Phosphorylation in sealed rod outer segments:effects of cyclic nucleotides. Biochemistry 23:515-521.

Sitaramayya, A. (1986) Rhodopsin kinase prepared from bovine rod disk membranes quenches light activation of cGMP phosphodiesterase in a reconstituted system. Biochemistry 25:5460-5468.

Sitaramayya, A. & Liebman, P.A. (1983a) Mechanism of ATP quench of phosphodiesterase activation in rod disc membranes. The Journal of Biological Chemistry 258:1205-1209.

Sitaramayya, A. & Liebman, P.A. (1983b) Phosphorylation of rhodopsin and quenching of cyclic GMP phosphodiesterase activation by ATP at weak bleaches. The Journal of Biological Chemistry 258:12106-12109.

Sneyd, J. & Tranchina, D. (1989) Phototransduction in cones: An inverse problem in enzyme kinetics. Bulletin of Mathematical Biology 51:749-784.

Steinberg, R.H., Fisher, S.K. & Anderson, D.H. (1980) Disc morphogenesis in vertebrate photoreceptors. The Journal of Comparative Neurology 190:501-518.

Stryer, L. (1991) Visual excitation and recovery. The Journal of Biological Chemistry 266:10711-10714.

Suh, K.H. & Hamm, H. (1988) Components I and II form a complex with the beta gamma subunits of G protein in frog rod outer segments. Investigative Ophthalmology and Visual Science 28:218.

Tamura, T., Nakatani, K. & Yau, K.-W. (1991) Calcium feedback and sensitivity regulation in primate rods. The Journal of General Physiology 98:95-130.

Ting, T.D. & Ho, Y.-K. (1991) Molecular mechanism of GTP hydrolysis by bovine transducin: Pre-steady-state kinetic analyses. Biochemistry 30:8996-9007.

Tranchina, D., Sneyd, J. & Cadenas, I.D. (1991) Light adaptation in turtle cones. Testing and analysis of a model for phototransduction. Biophysical Journal 60:217-237.

Tsuboi,S. Matsumoto,H., Jackson,K.W., Tsujimoto,K., Williams,T. & Yamazaki,A. (1994) Phosphorylation of an inhibitory subunit of cGMP phosphodiesterase in Rana catesbiana rod photoreceptors. I. Characterization of the phosphorylation. The Journal of Biological Chemistry 269:15016-15023.

Tsuboi,S. Matsumoto,H. & Yamazaki,A. (1994) Phosphorylation of an inhibitory subunit of cGMP phosphodiesterase in Rana catesbiana rod photoreceptors. II. A possible mechanism for the turnoff of cGMP phosphodiesterase without GTP hydrolysis. The Journal of Biological Chemistry 269:15024-15029.

Udovichenko, I.P., Cunnick, J., Gonzalez, K. & Takemoto, D.J. (1994) Functional effect of phosphorylation of the photoreceptor phosphodiesterase inhibitory subunit by protein kinase C. The Journal of Biological Chemistry 269:9850-9856.

Uhl, R., Wagner, R. & Ryba, N. (1990) Watching G proteins at work. Trends in Neuroscience 13:64-70.

Vuong, R.M., Chabre, M. & Stryer, L. (1984) Millisecond activation of transducin in the cyclic nucleotide cascade of vision. Nature 311:659-661.

Vuong, T.M. & Chabre, M. (1990) Subsecond deactivation of transducin by endogenous GTP hydrolysis. Nature 346:71-74.

Vuong, T.M. & Chabre, M. (1991) Deactivation kinetics of the transduction cascade of vision. Proceedings of the National Academy of Sciences USA 88:9813-9817.

Wagner, R., Ryba, N. & Uhl, R. (1988) Sub-second turnover of transducin GTPase in bovine rod outer segments. FEBS Letters 234:44-48.

Wagner, R., Ryba, N. & Uhl, R. (1989) Calcium regulates the rate of rhodopsin disactivation and the primary amplification step in visual transduction. FEBS Letters 242:249-254.

Wilden, U., Hall, S.W. & Kuhn, H. (1986) Phosphodiesterase activation by photoexcited rhodopsin is quenched when rhodopsin is phosphorylated and binds the intrinsic 48 KDa protein of rod outer segments. Proceedings of the National Academy of Sciences USA 83:1174-1178.

Wilden, U. & Kuhn, H. (1982) Light-dependent phosphorylation of rhodopsin:number of phosphorylation sites. Biochemistry 21:3014-3022.

Williams, D.S., Linberg, K.A., Vaughan, D.K., Fariss, R.N. & Fisher, S.K. (1988) Disruption of microfilament organization and deregulation of disk membrane morphogenesis by cytochalasin D in rod and cone photoreceptors. The Journal of Comparative Neurology 272:161-176.

Yamazaki, A., Hayashi, F., Tatsumi, M., Bitensky, M.W. & George, J.S. (1990) Interactions between the subunits of transducin and cyclic GMP phosphodiesterase in Rana catesbiana rod photoreceptors. The Journal of Biological Chemistry 265:11539-11548.

Yarfitz, S. & Hurley, J.B. (1994) Transduction mechanisms of vertebrate and invertebrate photoreceptors. The Journal of Biological Chemistry 269:14329-14332.

Yau, K.-W. (1994) Phototransduction mechanism in retinal rods and cones: The Friedenwald lecture. Investigative Ophthalmology and Visual Science 35:9-32.

Yau, K.-W. & Baylor, D.A. (1989) Cyclic GMP-activated conductance of retinal photoreceptor cells. Annual Reviews of Neuroscience 12:289-327.

Younger, J.P., McCarthy, S.T. & Owen, W.G. (1992) Modulation of cytosolic free calcium and changes in sensitivity and circulating current occur over the same range of steady-state adapting lights in rod photoreceptors. Investigative Ophthalmology and Visual Science 33:1104.