The Visual Pigments of Geckos and other Vertebrates: An Essay in Comparative Biology

Visual cells and visual pigments of the river lamprey revisited

Retinas of the river lamprey Lampetra fluviatilis were studied by microspectrophotometry, electroretinography and single-photoreceptor electrophysiology to reconcile the apparently contradictory conclusions on the nature of lamprey photoreceptor cells drawn in the early work by Govardovskii and Lychakov (J Comp Physiology A 154:279-286, 1984) and in recent studies. In agreement with recent works, we confirmed former identification of short photoreceptors as rods and of long photoreceptors as cones. In line with the results of 1984, we show that within a certain range of light intensities the lamprey retina exhibits "color discrimination". We found that the overlap of working intensity ranges of rods and cones is not a unique feature of lamprey short receptors, and suggest that rod-cone (possibly color) vision may be common among vertebrates. We show that the decay of meta-intermediates in lamprey cones occurs almost 100 times faster than in typical rod metarhodopsins. Rate of decay of metarhodopsins of lamprey rods take an intermediate position between typical rods and cones. This makes lamprey rhodopsin similar to transmuted cone visual pigment in "rods" of nocturnal geckos. We argue that defining various types of photoreceptors as simply "rods" and "cones" may be functionally correct, but neglects their genetic, biochemical and morphological features and evolutionary history.

Scene through the eyes of an apex predator: a comparative analysis of the shark visual system

The eyes of apex predators, such as the shark, have fascinated comparative visual neuroscientists for hundreds of years with respect to how they perceive the dark depths of their ocean realm or the visual scene in search of prey. As the earliest representatives of the first stage in the evolution of jawed vertebrates, sharks have an important role to play in our understanding of the evolution of the vertebrate eye, including that of humans. This comprehensive review covers the structure and function of all the major ocular components in sharks and how they are adapted to a range of underwater light environments. A comparative approach is used to identify: species-specific diversity in the perception of clear optical images; photoreception for various visual behaviours; the trade-off between image resolution and sensitivity; and visual processing under a range of levels of illumination. The application of this knowledge is also discussed with respect to the conservation of this important group of cartilaginous fishes.

G protein-coupled receptor rhodopsin: a prospectus

Rhodopsin is a retinal photoreceptor protein of bipartite structure consisting of the transmembrane protein opsin and a light-sensitive chromophore 11-cis-retinal, linked to opsin via a protonated Schiff base. Studies on rhodopsin have unveiled many structural and functional features that are common to a large and pharmacologically important group of proteins from the G protein-coupled receptor (GPCR) superfamily, of which rhodopsin is the best-studied member. In this work, we focus on structural features of rhodopsin as revealed by many biochemical and structural investigations. In particular, the high-resolution structure of bovine rhodopsin provides a template for understanding how GPCRs work. We describe the sensitivity and complexity of rhodopsin that lead to its important role in vision.

Distribution of rod- and cone-specific phosducins in retinas of non-mammalian vertebrates

In mammalian retinas, it has been believed that just one kind of phosducin (PD) commonly exists in both rods and cones. However, we have previously reported that there are rod- and cone-specific PDs (OlPD-R and OlPD-C) in medaka (Oryzias latipes) retina [FEBS Lett., 502, 117-121, 2001]. To clarify the distribution and evolution of these photoreceptor type-specific PDs, we investigated PDs of another teleost and a reptile. Immunohistochemical and Western blot analyses using anti-medaka PD antisera demonstrated that two kinds of PDs are expressed in zebrafish (Danio rerio) photoreceptor cells. Our study is suggestive that teleosts generally possess rod- and cone-specific PDs. We isolated a cDNA encoding putative PD (PmlPD) of a diurnal gecko (Phelsuma madagascariensis longinsulae). Because diurnal gecko possesses a pure-cone retina, it was expected that PmlPD would be expressed in cones. Molecular phylogenetic analysis demonstrated that PmlPD was more closely related to mammalian PDs than teleost cone-specific PDs, suggesting that the rod- and cone-specific subtype of teleost PDs have arisen after the teleost-tetrapod divergence.

Chloride effect on iodopsin studied by low-temperature visible and infrared spectroscopies

To investigate the chloride effect on the spectral properties of iodopsin, we have prepared an anion-free iodopsin (iodopsin.free) by extensive dialysis of an iodopsin sample against a buffer containing no chloride, and visible and infrared difference spectra between iodopsin.free and its photoproduct at 77 K were recorded. The absorption maximum of iodopsin.free in L-alpha-phosphatidylcholine liposomes was 528 nm, which was almost identical with that of the nitrate-bound form of iodopsin (526 nm, iodopsin.NO(3)), but 43 nm blue-shifted from that of the chloride-bound form of iodopsin (iodopsin.Cl). The iod/batho visible difference spectrum obtained from iodopsin.free was similar in shape to that from iodopsin.NO(3), but not to that from iodopsin.Cl. FTIR spectroscopy revealed that the chromophore vibrational bands and the peptide bonds of the original state in iodopsin.free were identical with those in iodopsin.NO(3) and were also similar to those in iodopsin.Cl except for the ethylenic vibrations of the chromophore. In contrast, those of the batho state in iodopsin.free were similar to those in iodopsin.NO(3) but considerably different from those in iodopsin.Cl. These results suggested that the binding of chloride but not nitrate induces a conformational change in the protein and that the chloride binding site is situated in a position where it directly interacts with the chromophore when the chromophore is photoisomerized. FTIR spectroscopy also revealed that one of the four water bands observed in the batho/iod spectrum of iodospin.Cl is absent in the spectra of iodopsin.free and iodopsin.NO(3). Thus, in contrast to nitrate, a lyotropic anion, chloride would bind to the binding site with water molecule(s) which could form a hydrogen-bonding network with amino acid residue(s) near the chromophore, thereby resulting in the red shift of the absorption maximum of iodopsin.

Evolution of visual pigments and related molecules

The molecular phylogenetic tree of vertebrate visual pigments, constructed on the basis of amino acid sequence identity, suggests that the visual pigments can be classified into five groups (L, ML, MS, S and Rh) and that their genes have evolved along these five gene lines. Goldfish has a UV-sensitive visual pigment (S group) localized in miniature single cone cells. Medaka has one type of rod cell containing rhodopsin (Rh group) and four types of cone cells, each of which contains a specific visual pigment with an absorption maximum that differs from those of the others. Frogs have a violet-sensitive visual pigment (S group) in small single cone cells and a blue-sensitive visual pigment (MS group) in green rod cells. Although nocturnal and diurnal geckos have rod- and cone-based retinas, respectively, they have phylogenetically closely related visual pigments. The pigments in each line may have restricted absorption maxima. We have cloned cDNAs encoding molecules involved in the phototransduction system of visual cells, such as phosphodiesterase, opsin kinase and arrestin. We then constructed phylogenetic trees of these molecules with the deduced amino acid sequences. The resulting phylogenetic trees show that these molecules are classified into two groups; one is expressed in cones and another in rods, suggesting that rods and cones contain homologous molecules with different amino acid sequences. These differences may result in the different light responses of rods and cones.

Chloride binding regulates the Schiff base pK in gecko P521 cone-type visual pigment

The binding of chloride is known to shift the absorption spectrum of most long-wavelength-absorbing cone-type visual pigments roughly 30 nm to the red. We determined that the chloride binding constant for this color shift in the gecko P521 visual pigment is 0.4 mM at pH 6.0. We found an additional effect of chloride on the P521 pigment: the apparent pKa of the Schiff base in P521 is greatly increased as the chloride concentration is increased. The apparent Schiff base pKa shifts from 8.4 for the chloride-free form to >10.4 for the chloride-bound form. We show that this shift is due to chloride binding to the pigment, not to the screening of the membrane surface charges by chloride ions. We also found that at high pH, the absorption maximum of the chloride-free pigment shifts from 495 to 475 nm. We suggest that the chloride-dependent shift of the apparent Schiff base pKa is due to the deprotonation of a residue in the chloride binding site with a pKa of ca. 8.5, roughly that of the Schiff base in the absence of chloride. The deprotonation of this site results in the formation of the 475 nm pigment and a 100-fold decrease in the pigment's ability to bind chloride. Increasing the concentration of chloride results in the stabilization of the protonated state of this residue in the chloride binding site and thus increased chloride binding with an accompanying increase in the Schiff base pK.

The unique lipid composition of gecko (Gekko Gekko) photoreceptor outer segment membranes

This study investigated the lipid and fatty acid composition of gecko photoreceptor outer segment membranes which contain the P521 cone-type pigment. The lipids of gecko photoreceptor outer segment membranes were first extracted and separated by thin layer chromatography (TLC) and then analyzed by gas chromatography (GC). Our results show that gecko photoreceptor outer segment membranes contain less phosphatidylethanolamine (PE) and more phosphatidylcholine (PC) and phosphatidylserine (PS) compared with those of bovine and frog. The content of the polyunsaturated fatty acid, docosahexaenoic acid (DHA), in PC and PS is also the highest yet reported (55 and 63%, respectively). These lipid differences may provide some insight into the specific lipid requirements of cone-type pigments.

Early photolysis intermediates of gecko and bovine artificial visual pigments

Nanosecond laser photolysis measurements were conducted on digitonin extracts of artificial pigments prepared from the cone-type visual pigment, P521, of the Tokay gecko (Gekko gekko) retina. Artificial pigments were prepared by regeneration of bleached gecko photoreceptor membranes with 9-cis-retinal, 9-cis-14-methylretinal, or 9-cis-alpha-retinal. Absorbance difference spectra were recorded at a sequence of time delays from 30 ns to 60 microseconds following excitation with a pulse of 477-nm actinic light. Global analysis showed the kinetic data for all three artificial gecko pigments to be best fit by two-exponential processes. These two-exponential decays correspond to similar decays observed after photolysis of P521 itself, with the first process being the decay of the equilibrated P521 Batho<-->P521 BSI mixture to P521 Lumi and the second process being the decay of P521 Lumi to P521 Meta I. In spite of its large blue shift relative to P521, iso-P521 displays a normal chloride depletion induced blue shift. Iso-P521's early intermediates up to Lumi were also blue-shifted, with the P521 Batho<-->P521 BSI equilibrated mixture being 15 nm blue-shifted and P521 Lumi being 8 nm blue-shifted relative to the intermediates formed after P521 photolysis. The blue shift associated with the iso-pigment is reduced or disappears entirely by P521 Meta I. Similar blue shifts were observed for the early intermediates observed after photolysis of bovine isorhodopsin, with the Lumi intermediate blue-shifted 5 nm compared to the Lumi intermediate formed after photolysis of bovine rhodopsin. These shifts indicate that a difference exists between the binding sites of 9- and 11-cis pigments which persists for microseconds at 20 degrees C.

Structure and photobleaching process of chicken iodopsin

Iodopsin, a dominant cone pigment in a chicken retina, has an absorption spectrum in longer wavelength region than rhodopsin. To account for this red-shift of iodopsin, we had proposed a structural model from retinal analogue experiments, in which iodopsin would have a relatively long distance between the protonated Schiff base nitrogen and the counterion. This was confirmed by a resonance Raman spectroscopy. The photochemical properties of iodopsin were studied and compared with those of rhodopsin, which revealed the following differences. The regeneration rate of iodopsin with 11-cis-retinal was 240 times faster than rhodopsin. Meta-iodopsin II, the signalling state of iodopsin, decayed about 100 times faster than meta-rhodopsin II. The Km value of meta-iodopsin II and rhodopsin kinase was lower than meta-rhodopsin II. These results are in consistent with rapid adaptation and low photosensitivity of cones relative to those of rods.

Presence and foveal enrichment of rod opsin in the "all cone" retina of the American chameleon

The retinal photoreceptors of the eye of the American chameleon, Anolis carolinensis, have been considered to be exclusively cones. Its retina is unusual for possessing two foveas (areas associated with heightened visual acuity), with the major, central fovea deeply incised and very densely packed with photoreceptors. Immunoblotting and light- and electron microscopic-immunocytochemistry, using several opsin monoclonal antibodies previously found specific for rods, demonstrated the presence and localization of this protein in the Anolis retina. This visual pigment appears sparsely in a subpopulation of photoreceptors in the periphery but overwhelmingly in the central fovea. Complementary results with cone-specific antibody and lectin binding corroborated this spatial organization. These results, as well as those with geckos, suggest that photoreceptor morphology is not an accurate guide among the lacertilians to visual pigment content, and that this phylogenetic grouping may constitute a crossroads in vertebrate photoreceptor evolution.

Identification of the Cl(-)-binding site in the human red and green color vision pigments

Chloride ions are known to bind and alter the absorption spectra of some but not all visual pigments. In this report, the human red and green color vision pigments are shown to bind Cl- and to undergo a large red shift in their absorption maxima. Mutation of 18 different positively charged amino acids in these pigments identified two residues, His197 and Lys200, in the Cl(-)-binding site. His197 and Lys200 are strictly conserved in all long-wavelength cone pigments but are absent in all rhodopsins and short-wavelength cone pigments. This fact suggests that the evolutionary branch of the long-wavelength pigments was established when an ancestral pigment acquired the ability to bind Cl- and, as a result, shift the absorption maximum to longer wavelengths.

The Gecko visual pigment: the dark exchange reaction and the effects of anions

A dark reaction is known to occur in retinal extracts of the gecko (Gekko gekko), in which the natural 11-cis-chromophore of the 521-pigment is apparently replaced by adding 9-cis-retinal to form the 9-cis-photopigment. With chloride-deficient extracts the reaction involves some 70% of the 521-pigment. Anions like nitrate, azide, thiocyanate and cyanate that shift the spectrum toward the blue do not affect this 70% exchange. Anions like fluoride, iodide and sulfate likewise do not alter this magnitude of reaction. In contrast, chloride and bromide that induce a bathochromic spectral shift lead to a decrease in this dark replacement of the 11-cis chromophore. This protection is similar to the action of these two anions in antagonizing the pigment loss by NH2OH and by temperature, both occurring in the dark. Apparently, chloride and bromide alter the opsin conformation so as to stabilize and/or protect the Schiff's base linkage but nitrate, azide, thiocyanate and cyanate act at a different opsin site or by a different mechanism.

The spectral properties and photosensitivities of analogue photopigments regenerated with 10- and 14-substituted retinal analogues

Analogues of 11-cis- and 9-cis-retinal with substitutions at positions 10 and 14 were used to regenerate analogue photopigments with two opsins: that of the transmuted (cone-like) 521-pigment of Gekko gekko and that of the rhodopsin of Porichthys notatus. The spectral absorbances and photosensitivities of the regenerated photopigments were determined and compared, first, between the two systems of analogue photopigments, and second, in the responses to the two opsins. Unlike the 10-fluoropigments, the comparable 14-compounds were significantly red-shifted by 19-30 nm and their sensitivity to light was similar to that of the parent 11-cis- and 9-cis-pigments. These were the results for both analogue pigments. In contrast, the 10-pigments were spectrally located close to the wavelengths of the parent compounds and the photosensitivity was significantly reduced, especially in the case of the 9-cis-analogues. Evidence was obtained for a steric hindrance effect at position 14, for no regeneration was obtained when methyl or ethyl groups were at this carbon. In the 10-substituted retinals, steric hindrance was noted only for the gecko; only the fluorosubstituted, but not the chloro-, the methyl- or the ethyl-substituted, retinals reacted. With the fish opsin, pigments were regenerated with all but the ethyl-substituted retinal. The gecko opsin appears to have a more restricted binding site. Another feature of the gecko was related to the chloride bathochromic and hyperchromic effects, in which the 521-pigment prepared in a chloride-deficient state has a blue-shifted spectrum compared with the spectrum obtained after the addition of chloride, and its extinction is raised by the addition of chloride to give a mean ratio of 1.23 for the two extinctions, one with, the other without, added chloride. The 11-cis-10-F-analogue pigment gave both chloride effects and the hyperchromic ratio was the same as that recorded for the native visual pigment. In contrast, the pigment formed with 11-cis-14-F-retinal gave a hyperchromic ratio significantly greater than 1.23. A similar contrast in the responses to chloride was obtained with the analogue photopigments regenerated with the 9-cis-10-F- and 9-cis-14-F-chromophores. This difference between the two systems is interpreted as the result of a specific configurational feature of the gecko opsin when in the chloride-deficient state that is relevant to the binding of the retinal analogue.(ABSTRACT TRUNCATED AT 400 WORDS)

Serotonin and dopamine in the retina of a lizard

The objective of this paper is to report the presence and localization of serotonin and dopamine in the retina of the lizard Uta stansburiana. High performance liquid chromatography and electrochemical detection were used to identify and quantitate the two amines. Both compounds are present as endogenous molecules in this retina and are found in concentrations similar to those reported in other non-mammalian retinas. The same methods were employed to confirm, in the isolated retina, the synthesis of serotonin from precursor, tryptophan. Immunocytochemical methods were used to localize, in the neural retina, serotonin and the rate-limiting enzyme of dopamine synthesis, tyrosine hydroxylase. Serotonin immunoreactivity was observed in bistratified amacrine cells (ca. 7 micron dia.) with processes ramifying in sublayers 1, 4, and 5 of the inner plexiform layer. Immunoreactivity to tyrosine hydroxylase was observed in a different population of bistratified amacrine cells (ca. 11 micron dia.) that had processes ramifying in sublayers 1 and 5 (and perhaps 3) of the inner plexiform layer. The enzymes for further metabolism of dopamine were not found in the retina of this lizard by immunocytochemical methods. The results of this research suggest that only single classes of serotoninergic and dopaminergic neurons are present in the retina of U. stansburiana. This retina might, therefore be an appropriate place in which to investigate the functioning of these amines in visual information processing.

Photosensitivity of 10-substituted visual pigment analogues: detection of a specific secondary opsin-retinal interaction

The photosensitivities of the bovine rhodopsin and gecko pigment 521 analogues regenerated from C-10-substituted analogues of 11-cis- and 9-cis-retinals were determined by two different methods. A similar reactivity trend was noted for both pigment systems as revealed in the photosensitivity of the gecko pigments and relative quantum yields of the bovine analogues. The 10-fluoro-11-cis photopigments had a photosensitivity less than, but approaching, that of the native (11-cis) visual pigment while the 10-fluoro-9-cis photopigments had a much lower photosensitivity than the parent 9-cis regenerated pigment. The results are interpreted in terms of recently described models of rhodopsin architecture and of the primary molecular reaction of visual pigments to light. The unusually low photoreactivity of the 10-fluoro-9-cis pigment molecule is viewed as the result of a regiospecific hydrogen-bonding interaction of the electronegative fluorine atom to the opsin.

The gecko visual pigment: its photosensitivity and the effects of chloride and nitrate ions

By use of the method of photometric curves, the photosensitivity of the major and ion-sensitive pigment of Gekko gekko has been determined and compared with that of rhodopsins of the frog (Rana pipiens) and of the fish (Porichthys notatus). In the presence of Cl- (or Br-), the gecko pigment has the same photosensitivity as the other A1 rod pigments, but unlike these, the addition of NH2OH does not lead to a Dartnall effect, i.e. an enhancement in the measured rate of photic bleaching. This is because the gecko pigment has no meta-III intermediate. In the Cl- -deficient state the gecko pigment has a photosensitivity 0.8 times that of the Cl- -provided system. The increase in photosensitivity brought on by Cl- is quantitatively accounted for by the Cl- -induced hyperchromic effect. The addition of NH2OH to the system without added Cl- leads to a small increase in measured rate of photic bleaching with an apparent 13% increment in photosensitivity. This is not a classical Dartnall effect for here again no meta-III is involved. The possibility is raised of an additional, yet undiscovered, action of NH2OH on the opsin moiety. Nitrate ions (NO3-) are known to produce an increase in extinction coefficient similar to that of Cl- and a hypochromic shift in the spectral absorbance. Despite the hyperchromic action, NO3- produces a reduction in the measured rate of photic bleaching, an effect explained by the appearance of a meta-III type intermediate absorbing at about 470 nm. While Cl- is able to antagonize the NO3- -induced hypochromic shift, it is unable to reverse the NO3- -induction of meta-III. This, along with other differences in responses of the gecko pigment to these two ions, suggests that Cl- and NO3- act at two different sites and produce unique conformational changes in the protein molecule.