Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Optical and Molecular Design of Rods WILLIAM H. MILLER INTRODUCTION Thank you, Bob Shapley, for such a kind introduction. I greatly appreciate it. In trying to strike a keynote both melodious and strong enough to resonate with this audience, I think of Franz Boll, who proposed the photochemical theory of vision. Although he died of tuberculosis at the age of 30, before he had an opportunity to exploit his great discovery of visual purple, there is no doubt that he appreciated its potential as discussed in Hubbard's (1977) preface to her translation of his paper. Not only did Boll propose the photochemical theory of phototrans- duction, but he also proposed a photophysical theory, which has found application in explaining the color of another of his discoveries, the green rod. Because I have been interested in transduction and the green rod, I want to share some of my thoughts on both of these problems with you. GREEN ROD Boll's green rod represents a design problem dealing with both visual pigments and image sampling. Neither the rod's color nor its role in sam- pling is completely understood. Alla n Snyder and I have proposed that the green rod myoid is a wavelength filter (Miller and Snyder, 1972~. Reuter and Virtanen (1972) showed that the green rod is the short-wavelength receptor for a color opponent process at photopic and mesopic light inten- sities. In contrast to the thin rods with pooled responses of many species, Jagger (1988) suggested that amphibian fat rods may individually sample 2
OPTICAL AND MOLECULAR DESIGN OF RODS 3 FIGURE 1 End-on appearance of pink and green rods. Pink rods are bleached; green rods retain their green color SOURCE: Boll (1877~. the eye's image. The green rod may function with both the rod and cone systems for image sampling and color vision. Green Color Not Caused by Visual Pigment Alone Boll was the first to describe the green rod. He proposed a photo- physical theory of vision in addition to his photochemical theory because he was uncertain as to the origins of rod colors (Boll, 1877~: As against this photochemical theory of the nature of visual red and of the light sense, another formulation is possible which in contrast might be called the photophysical theory. This theory assumes that there is no special pigment that permeates the rod stroma, but attributes the red color of the rods to a purely physical phenomenon, the optical effect of stacking of intrinsically colorless platelets. Accordingly visual red would belong to the category of interference phenomena and more specifically to the narrower class of the so-called colon of thin films. The photophysical theory of vision was not taken seriously after Kuhne's characterizations of rhodopsin. Nevertheless, this concept was plausible at the time. The dimensions of the rod platelets would not be known for 75 years, and in fact such platelets are found to produce inter- ference colors in the optical trains of both invertebrate and vertebrate eyes. More to the point, a photophysical theory has merit when applied to the case of the amphibian green rod. Figure 1 is a black and white reproduction of part of Boll's 1877 color plate showing the frog rod mosaic with its pink and green rods after illumination with short-wavelength light. The dark spots are green in the original color plate, and the light spots representing the bleached pink rods are colorless. Figure 2 is a scanning electron micrograph of part of the photoreceptor layer of a preparation of an amphibian retina similar to that used by Boll. Although the green rod has a shorter outer segment than the pink rod, it nevertheless retains its green color after the pigment of the pink rod has
WILLL4M H. MILLER FIGURE 2 Scanning electron micrograph of amphibian photoreceptor layer showing a few single cones, many pink rods, and one green rod with its 1-micron diameter myoid. NOTE: Arrow indicates the external limiting membrane been bleached, as shown in Boll's figure. This was confirmed by Liebman and Entine (1968~: It is of some interest that our green rods show no light absorption in the red- orange region that would explain results of Denton and Willie who found such absorption in end-on measurements using photographic densitometty (Demon and Wyllie, 1955~. Nor, after bleaching, was an increased absorption found in the green region of the spectrum....We can therefore state categorically that the locus of the green and red absorbing entities is not in pigment form in rod outer segments. Similarly, a year earlier Dartnall (1967) also reached the conclusion that the yellow-red absorption may "even possibly be a structural color."
OPIICAL AND MOLECULAR DESIGN OF RODS s Myoids Are Waveguides Miller and Snyder (1972) proposed that the myoid's waveguide prop- erties in combination with the rod's 433-nm pigment could account for the green color. The assumption underlying this hypothesis is that the parts of the photoreceptor cells that are located sclerad to the external limiting membrane (horizontal arrow, Fig. 2) are light pipes. The green rod's aper- ture is its myoid near the external limiting membrane. Light that reaches the outer segment must first pass through the myoid. The physical prop- erties of the myoid, its diameter and the difference between its refractive index and that of its surround, should determine the intensity and spectral distribution of the light that reaches the green-rod visual pigment. The green rod's unbleached end-on color in white light results from absorption in the blue by its 433-nm pigment and "absorption" of yellow and red by an unknown structure. The myoid could be that unknown structure. Yellow-red light would be lost not by absorption but by not being propagated in the waveguide by total internal reflection. The light propagated in waveguides shows intensity patterns called modes, when the guide is viewed in cross section. For each mode the fraction of modal energy propagated inside the guide is a monotonic function of the ratio of guide diameter (d) to wavelength (~) and of the difference in refractive index between the guide (ins ~ and its cladding (nay. The smaller d is relative to ~ and the smaller non2, the less longer-wavelength light will reach the outer segment. The fraction of light energy inside the guide, '7, is equal to 1 - (7rdul) (o2 02) (TIC) where be, the complement of the critical angle, is equal to 1 - (n2/n~2 and ~ is the angle of incidence, which is a function of the mode number (equation from Snyder, 1975:p.44~. The fraction of light (~) that would be inside a myoid of 2 Em (Figure 3 curve c), 1 Em (curve a), 0.8 Em (curve b) and 0.5 Em diameter (curve d) is plotted as a function of wavelength based on the two lowest order modes in Figure 3 from Miller and Snyder (1972~. Curve a of Figure 3 shows that the dark-adapted myoid with a diameter of 1 Em diameter would propagate yellow-red wavelengths poorly in com- parison with shorter wavelengths. When viewed end on, the combination of decreased propagation in the yellow-red and absorption in the blue by the 432 rim pigment could cause the green appearance of the dark-adapted green rod. According to Arey (1916), the green rod myoid becomes slightly more tenuous when light-adapted. A 0.8 Em diameter would shift the curve to the left as shown by curve b, to mimic what was interpreted by Denton and Wyllie (1955) as an increased green absorption in the light-adapted
6 WILLIAM H. MILLER 10 . _ ~^ 0 8 . _ 3 06 - ° 0-4 o 0 02 . _ t ~ o 1 2'Lm ~ 05f~m ,Lm fib 1 1 300 350 400 450 500 Green 550 600 650 U ~ Yellow Red FIGURE 3 Theoretical effect of myoid diameter on fraction of light inside the myoid as a function of wavelength for the two lowest-order modes. SOURCE: Miller and Snyder (1972). end-on preparation. An even smaller diameter myoid, as illustrated in curse d, would tend to cut off much of the light in the visible spectrum. Green Rod Function Green Rods May be Color Receptors If we accept that the green myoid in combination with the outer- segment 433-nm pigment can explain the rod's end-on color, that still leaves the problem of its function. Reuter and Virtanen (1972) have said: Our experiments suggest the existence in the frog retina of separate channels for green rod responses. When during dark adaptation the red rods begin to contribute they enter the on-off and off channels of the class 1, 2 and 4 cells, that is the cone but not the green rod channels. This makes possible color vision in the mesopic state. The red rods simply contribute to the cone side of the opponent color system. This arrangement, together with the fact that the cone channels dominate and are sufficient for eliciting all the Maturanna et al. characteristics, guarantees the invariance of coding of brightness pattern information during the shift from photopic to scotopic vision. The main role of the green rod channels seems to be to "add color to the system."
OPIICAL AND MOLECULAR DESIGN OF RODS TABLE 1 Amphibian Rod and Cone Parameters 7 Green Rod Pink Rod Single Cone Outer-segment diameters) Dark-adapted entrance aperture (myoid at external limiting memberane) 4,um Plum 2.5~n =0.8pm2 ~28pm2 Outer-segment optical density, peak absorbance(a) 0.68 0.82 0.26 Relative anatomical sensiiiviy (absorbance x capture area) Sensitivity at peak wavelength (mV/quantum)(b~) Nyquist frequency based on posterior nodal distance = 5 mm (c/deg)(d) 2 0.6 23 ~10 =10 7 ~S[lm2 Qight adapted) 1 1 ? (a)Lieb~nan and Entine (1968) (Rana) (b)Fain and Dowling (1973) based on intracellular recordings in intact retina, which may exaggerate sensitivity of rods because of coupling. (C7Reuter and Virtanen (1972) sensitivity at peak wavelength measured at ganglion cell. (Jagger (1988) (Bunko). Green rods comprise 11% of total in plate la, ref.2; Nyquist frequency calculated assuming pink rod number density = 18000/mm2 and green rod number density = 1800/mm2. Are Individual Fat Rods Samplers? The rods may provide spatial vision matched to the optics under mesopic conditions, according to Jagger (1988~. Jagger argues that the rods could effectively sample the 6 cycles per degree image quality found over most of the eye's visual field. Some amphibian rod and cone parameters drawn from various sources are shown in Table 1. The green rod appears to be more sensitive than the single cone and may function as a color receptor. Comparison of the image sampling properties based on the information in the Table 1 is difficult because of the wide range of eye sizes among amphibian species. Using a 5-mm posterior nodal distance (PND) eye as a sample eye size, the green rods by themselves would not be useful for spatial resolution because they could support only a Nyquist frequency of 2 cycles per degree. It seems likely, therefore, that the green and pink rods function as a unit for color and spatial vision based on a flat 6 cycles per degree optical quality over most of the retina. There would be some undersampling (Snyder et al., 1986) by the ganglion cells in the region of peak density where the potential maximum acuity would be 4.4 cycles per degree, and there would be less than 3 rods per ganglion cell away from the visual streak. Nevertheless, the theory that
8 WILLIAM H. MILLER the fat amphibian rods individually sample the retinal image is plausible, considering that there is much image processing prior to the ganglion cells in the amphibian retina and there is an early optokinetic acuity measure of 4.4 cycles per degree for Rana (reviewed in Jagger, 1988~. Rods and cones might be expected to have small diameters not only because separate optical channels should be as small as possible, consistent with the diffraction limited image, to maximize resolution but also to facilitate temporal resolution by limiting diffusion transit times. Rod outer segments of many animals ranging from fish to mammals are about 2 Em in diameter. Those of amphibians are 6 Em or larger. A design by which amphibian fat rods may provide the same effective resolution over the entire visual field in amphibians contrasts with that of many retinas with thin rods in which the rod responses are pooled to provide increased sensitivity at the expense of resolution. The green rods may have evolved to contribute to both color and form vision. TRANSDUCTION For the study of transduction that Boll initiated, it is now generally accepted that the rising phase of the vertebrate photoreceptor response is mediated by activation of phosphodiesterase (PDE) and that recovery is mediated at least in part by activation of guanylate cyclase (cyclase). But there is controversy regarding whether inhibition of PDE contributes to the initial phase of recovery. Hiroaki Kondo and I (Kondo and Miller, 1988) suggested that the initial phase of recovery is driven by increased rather than decreased PDE activity. Increased PDE activity may lead to decreased [Caii and activation of cyclase to both truncate the response and cause the smaller amplitude and faster recovering light-adapted response. Photolyzed rhodopsin activates a cyclic nucleotide enzymatic cascade. The concentration of cyclic GMP is reduced causing decreased plasma- membrane conductance (Fesenko et al., 1985) and the decrease in dark current that is the response to illumination (see Pugh, in this volume; Stryer, 1986~. Details of the recovery process and light-adapted response are still not well understood. That calcium plays a role is generally accepted: cytoplasmic calcium concentration (jCaii) falls during the light response and that fall in [Caii is thought to either inhibit phosphodiesterase activity, activate guanylate cyclase activity, or both, to restore the concentration of cyclic GMP. Is Response Recovery Mediated Initially by Cyclase Activation? The development of the concept that recovery is mediated by activation of cyclase rests on the synthesis of two experimental facts: the finding of
OPTICAL AND MOLECULAR DESIGN OF RODS 9 McNaughton et al. (1986) that [Call is decreased by illumination and the findings of Lolley and Racz (1982), Pepe et al (1986) and Koch and Stryer (1988) that very low [Ca~i activates cyclase. Hodgkin et al. (1985) combined the first hints of these findings with their observation that calcium does not close channels as rapidly as light: Another possible explanation of the failure of calcium ions to close channels rapidly is that calcium might inhibit the production of a substance which keeps channels open rather than interacting directly with the channels. What could that substance be? On one hand, Cyclic GMP is a candidate based on evidence that that substance increases conductance of an excised patch of ROS plasma membrane (Fesenko et al. 1985~. On the other hand, though cyclic GMP increases channel conductance, whether that is because it binds the channel directly is not known, since the patch contains substances other than the ROS plasma membrane (Kolesnikov et al., 1987~. Nevertheless, measurements of [Ca~i have been continually refined. Dim illumination takes the dark resting level from 220 nM to ~140 nM (Ratto et al., 1988~; strong illumination has been calculated to give a minimum value of 60 nM (Miller and Korenbrot, 1987~. At the same time the evidence that cyclase is activated by nM levels of calcium has become convincing. Lolley and Racz (1982) were the first to use calcium at nM concentrations. Pepe et al. (1986) were able to duplicate Lolley and Racz's result on rat rods in dim red light, but only when their homogenized toad rods were illuminated. On the other hand, Koch and Stryer (1988) found that the activity of unilluminated bovine ROS cyclase was regulated by a highly cooperative calcium-dependent molecular switch (not calmodulin). Guanylate cyclase was activated up to 20-fold when [Ca] was reduced from 200 to 50 nM from a basal rate of 7 ~uM cyclic GMP/sec/ROS to 100 EM cyclic GMP/sec/ROS. Since [Caii may decrease from about 200 to 60 EM as a result of illumination, Koch and Stryer's evidence leads to the suggestion that the free concentration of cyclic GMP decreases only transiently in response to a flash, because low Ca rapidly triggers restoration of cyclic GMP levels. Pugh (in this volume) recounts other in press evidence to the same effect. Disagreement Regarding Role of PDE Inactivation While there is agreement that low [Caii aids recovery by activating cyclase, there is still no agreement as to whether the low [Ca]' also inhibits PDE. There is conflicting biochemical evidence regarding the rate of PDE inactivation. Recovery of PDE to its basal level is controlled by the rate of hydrolysis of GTP bound to alpha-transducin, Ta. On the one hand, the rate constant for transducin's GAP hydrolysis was measured to be 0.02 see-i (Bennett and Dupont, 1985), while other measurements show a
10 WILLIAM H. MILLER half life for Ta-GTP of 3 sec at 23° for cattle rods (Vuong and Chabre, 1988~. Decreased PDE activity would of course aid recovery of cyclic GMP levels near the plasma-membrane channel. In this vein, Pierre et al. (1986) conclude: The results are consistent with the idea that the rapid drop in Cai, which has recently been shown to accompany the light response, is involved in terminating the light response, and that Cai is thereby involved in setting the operating point and sensitivity of phototransduction. From the comparison with other work we infer that Cai appears to act, at least in part, by means of control of cGMP phosphodiesterase activity.... Hence we conclude that elevated Cai leads to enhanced activity of PDEase. An alternative hypothesis holds that lowered [Caii is involved in ini- tially terminating the light response not by inactivating PDE as suggested by Torre et al. but rather by activating cyclase. Light Response Recovery Because of PDE Activation? Uniform pC cyclic GMP injections into ROS at the resting potential cause cyclic GMP depolarizing responses of uniform amplitude and du- ration, as if the cyclic GMP responses provide an index of PDE activity. Corresponding with PDE activation by light, cyclic GMP responses are antagonized in a graded manner with increasing intensities of illumination (Miller, 1982~. Uniform 5 pC cyclic GMP pulses were delivered before during and after a 0.1 sec bright light flash in the experiment shown in Figure 4A. The cyclic GMP pulses are numbered 1-10. Each downward spike was an injection artifact. The numbered upward depolarizations were the cyclic GMP responses. The increased PDE activity that mediates the receptor potential was sufficient to reduce the amplitude and speed the recovery of cyclic GMP response number 2 in comparison with that of number 1. Although the receptor potential recovered rapidly, PDE activity (as reflected in the amplitude and recovery rate of the cyclic GMP response) continued to increase and did not recover to the preflash levels until more than a minute after the receptor potential recovered. The receptor potential recovered rapidly despite apparently high PDE activity. The light flash intensity was reduced by half and the flash duration was lengthened in Figure 4B. The recovery rate of the cyclic GMP responses was more nearly matched to that of the receptor potential. Finally in Figure 4C, in which the light flash was a log unit dimmer than that of Figure 4B, the recovery rate of the receptor potential matches the recovery rate of the cyclic GMP response (and presumably PDE activity). The initial very rapid recover from a bright flash occurs despite an apparent high rate of PDE activity, which implies that the concentration of cyclic GMP must be restored by cyclase activation rather than PDE
OPTICAL AND MOLECULAR DESIGN OF RODS 1 8 9 10 A it. ,-4 1c~ uni as 1 ~ 1 1 1 - ~ 1 Pi 1 B . |10mV 10s ffl<Trr 1 ~ ~ AFT I T T! c L~ -4.3 LOG UNITS -5.3 ~0G UNITS 11 Car \~: as'; .o ~ P. ~ _ a) C -a FIGURE 4 Voltage recordings from a single rod of the isolated frog retina for van- ous intensities and durations of illumination. The S-pC injections of cyclic GMP are superimposed on the traces as a measure of PDE activity. "Down" on the signal trace indicates injection; "up" signals light. Receptor potential is a hyperpolanzation indicated by downward movement of the recording trace. Downward spikes are injection artifacts; cyclic GMP responses are up. inactivation. A similar conclusion follows from the experiments of Ames et al. (1986) showing that cyclic GMP metabolic turnover increases with stimulus intensity (until saturation). Ames et al. (1986) found that cyclic GMP metabolic flux increased about 4.5-fold for a 3-log unit increase in light intensity with little change in measurable cyclic GMP levels. The dimmest intensity caused an estimated 30 photoisomerizations/ROS/sec to turn over about 104 cyclic GMP molecules/Ah*; the brightest continuous illumination turned over 573 cyclic GMP's/Rh*/sec for a total of 4 x 106 cyclic GMP's/ROS. Koch and Stryer's maximal cyclic GMP synthesis of 100 M/ROS/sec (on bovine in vitro rather than rabbit in vivo) corresponds to 5 x 107 cyclic GMP's/ROS/sec based on a ROS cytoplasmic volume of 10-~2 liters. The experiments in Figure 4 showing recovery despite high PDE activity, those of Ames et al. (1986) showing increased metabolic flux of cyclic GMP with increasing light stimulus intensity and of Koch and Stryer (1988) showing a steep Ca dependence of cyclase are most easily
12 pA WILLIAM H. MILLER ~ . c o r1 t ~ o 1 \ ~ ~ ~~ it= 3 FIGURE 5 Suction current responses to light flash before and after application of p[NH]ppG to rod outer segment by whole-cell patch. Response after ptNH]ppG is scaled up by factor of 2.17 for normalization. Response after p[NH]ppG is truncated presumably by cyclase activation as indicated in Figure 6. explained by a model in which the initial stage of recovery is mediated by an increase rather than a decrease in PDE activity. We obtained additional support for this hypothesis by testing whether the response to a light flash can recover when PDE activated by the flash is inhibited from deactivation by the introduction of the nonhydrolyzable analog of GTP, guanyl-5'-yl imidodiphosphate, p[NH]ppG, into the outer segment using the whole-cell patch clamp technique (Kondo and Miller, 1988~. The p[NH]ppG both reduces the amplitude of the response and hastens the initial segment of recovery (Figure 5~. A previous experiment in which a hydrolysis resistant analog of GTP was patched into the outer segment gave a different result. There was no response recovery when 500 ,uM GTP-^y-S was applied to the isolated outer segment (Sather and Detwiler, 1987~. The difference between these two results is difficult to explain in the absence of information about the Km's of the various analogs and of GTP for cyclase, but a preliminary report of 1-2 mM ptNHippG patched into the inner segment of the isolated rod (Lamb and Mathews, 1986) appears to show results similar to those of the whole cell rather than the isolated outer segment. The effects of piNH]ppG may be explained by a model in which activated transducin leads to a new steady state of lowered dark current and hence sensitivity. An added test flash rapidly leads to still lower cytoplasmic calcium concentration, and because of the steepness of the calcium dependence of cyclase, the concentration of cytoplasmic cyclic GMP is driven rapidly higher to truncate the response. The later and slower component of recovery would be mediated by Ta.GTPase. This analogy between the effects of ptNH]ppG and the effects of light adaptation leads to the suggestion (Kondo and Miller, 1988) that transducin activated by photolyzed rhodopsin is the signal that controls the initial rod response recovery following a bright flash as well as the rapid initial recovery and
OPTICAL AND MOLECULAR DESIGN OF RODS g-cyclase. to [Ca i] ~ ROS inward Ca current GTP ~ ~ 5'GMP - ~ - ~ PDE* 13 / cyclic GMP T*` R* FIGURE 6 Schematic of hypothesized molecular mechanism of initial response recovery and of light adaptation of rod photoreceptor. Light revs up PDE~dase cycle, causing it to race. Light adaptation reduces the ROS inward calcium current and [Caii, which activates g-pyclase to rapidly increase cyclic GMP levels, which truncates response. reduced sensitivity of the light-adapted response by means of acceleration of the PDE-cyclase cycle as indicated in Figure 6. CONCLUSION Remarquez bien que les nez ont ete fan's pour porter des lunettes, aussi nous-avons des lunettes. Candide, Ch. 1:38-39 The speculators continue to give rise to suggestions for optical and physiological photoreceptor functions following the good example of Franz Boll, who was the first to practice this art in visual science. The suggestions presented at this symposium are grist for the mill. Because observation
14 WILLIAM H. MILLER continually improves, these speculations on the path to knowledge can never be accepted as final. Getting there is all the fun. ACKNOWLEDGMENTS This work was supported mainly by National Institutes of Health grant EY03196 and in part by NIH Vision Center grant EY00785, Research to Prevent Blindness, Inc., and the Connecticut Lions Eye Research Founda- tion. Note added in proof: There have been significant developments since this manuscript was written. 1. Additional evidence that calcium may mediate the light-adapted response was published; reviewed by: E. Pugh and J. Altman (1988) A role for calcium in adaptation. Nature 334: 1~ 17. 2. Studies showing cyclic GMP Sated channel activity lent additional support for the theory that cyclic GMP may mediate transduction by directly binding the light-sensitive channel: J.W. Clack and P.J. Stein (1988) Opsin exhibits cGMP-activated single-channel activity. Proc. Natl. Acad. Sci USA 85:9806 9810 and U.B. Kaupp, T. Niidome, T. Knave, S. Ibrada, W. Bonigk, W. Stuhmer, N.J. Cook, K Kangawa, H. Matsuo, T. Hirose, T. Miyata, and S. Numa (1989) Primary structure and functional expression from complementary DNA of the rod photoreceptor cyclic GMP-gated channel. Nature 343:762-766. REFERENCES Ames, A. III, T.F. Walseth, R.A. Heyman, M. Barad, R.M. Graeff and N.D. Goldberg 1986 Light-induced increases in cGMP metabolic flux correspond with electrical responses of photoreceptors. The Joumal of Biological Chemisay 261:13034- 13042. Bennett, N., and Y. Dupont 1985 The G-protein of retinal rod outer segments (Transducin). The Journal of Biological Chemisay 26~.4156-4168. Boll, F. 1877 On the anatomy and physiology of the retina. (Translated by Ruth Hubbard with the help of Helene Hoffmann). Vision Research 17:1253-1266. Dartnall, H.J.A. 1967 The visual pigment of the green rods. Vision Research 7:1-16. Denton, E.J., and J.H. Wyllie 1955 Study of the photosensitive pigments of the pink and green rods of the frog. Journal of Physiology (London) 127:81-89. Fain, G.L., and J.E. Dowling 1973 Intracellular recordings from single rods and cones in the mudpuppy retina. Science 180:1178-1180. Fesenko, E.E., S.S., Kolesnikov, and A.L Lyubarsky 1985 Induction by cyclic GMP of cationic conductance in plasma membrane of retinal rod outer segment. Nature 313:310-313.
OPTICAL AND MOLECULAR DESIGN OF RODS 15 Hubbard, R. 1977 Preface to the English translations of Boll's On the anatomy md physiology of the retina and of Kuhne's Chemical processes in the retina. Vision Reseach 17:1247-1248 Hodgkin, A.L., P.A. McNaughton, and B.J. Nunn 1985 The ionic selectivity and calcium dependence of the light-sensitive pathway in toad rods. Joumal ofPhysiology (London) 358:447468 Jagger, W.S. 1988 Optical quality of the eye of the cane toad, Bufo marinas. Vision Research 28:105-114. Koch, K-W. and L. Stryer 1988 Highly cooperative feedback control of retinal guanylate cyclase by calcium ion Nature 334:64-66. Kolesnikov, S.S., A.B. Jainazarov, and E.E. Fesenko 1987 Time-dependent cGMP-activated conductance of detached patches of ROS plasma membrane. FEBS Letters 222:37-41. Kondo, H., and W.H. Miller 1988 Rod light adaptation may be mediated by acceleration of the phosphodiester- ase-guanylate cydase cycle. Proceedings of the National Academy of Sciences, USA 85:1322-1326. Lamb, T.D., and H.R. Mathews 1986 Incorporation of hydrolysis-resistant analogues of GTP into the rod photore- ceptors isolated from the tiger salamander. Joumal of Physiology (LondonJ 381:58p. Liebman, P.A., and G. Entine 1968 Visual Pigments of frog and tadpole (`Rana pipiens). Vision Research 8:761- 775. Lolley, R.N., and E. Racz 1982 Calcium modulation of cyclic GMP synthesis in rat visual cells. Vision Research 22:1481-1486. McNaughton, P.N, Lo Cervetto, and B.J. Nunn 1986 Measurment of the intracellular free calcium concentration in salamander rods. Nature 322:261-263. Miller, D.L, and J.I. Korenbrot 1987 Kinetics of light-dependent Ca fluxes across the plasma membrane of rod outer segments. Joumal of General Physiology 90:397-425. Miller, W.H. 1982 Physiological evidence that light-mediated decrease in cyclic GMP is an in- termedia~y process in retinal rod transduction. Joumal of General Physiology 80:103-123. Miller, W.H., and A.W. Snyder 1972 Optical function of myoids. Vision Research 12:1841-1848. Pepe, I.M., I. Panfoli, and C. Cugnoli 1986 Guanylate cyclase in rod outer segments of the toad retina. FEBS Letters 203:73-76. Ratto, G.M., R. Payne, W.G. Owen, and R.Y. Tsien 1988 Cytosolic free [Ca2+] of rods in the bullfrog retina measured with FIJRA2. Biophysical Joumal 53:473a. Reuter, T., and K. Virtanen 1972 Border and colour coding in the retina of the frog. Nature 239:260-263.
16 W7LLLAM [I. MILLER Sather, W.A., and P.B. Detwiler 1987 Intracellular biochemical manipulation of phototransduction in detached rod outer segments. Proceedings of the National Academy of Sciences, USA 84:9290-9294. Snyder, JAW. 1975 Photoreceptor optics-theoretical principles. In "Photoreceptor Optics," A.W. Snyder and R. Menzel, eds. Berlin: Springer-Verlag. Snyder, JAW., T.R.J. Bossomaier, and A. Hughes 1986 Optical image quality and the cone mosaic. Science 231:499-501. Stryer, L 1986 Cyclic GMP cascade of vision. Annual Reviews of Neuroscience 9:87-119. Torre, V., H.R. Matthews, and T.D. Lamb 1986 Role of calcium in regulating the cyclic GMP cascade of phototransduction in retinal rods. Proceedings of the National Academy of Sciences, USA 83:7109-7113. Vuong, T.M., and M. Chabre 1988 As measured by fast microcalorimet~y, bound GTP in transducin is hydrolysed within 3 seconds in cattle rods at 23°C. Biophysical Joumal 53:472a.
