Cephalochordate melanopsin: evolutionary linkage between invertebrate visual cells and vertebrate photosensitive retinal ganglion cells

Animal photoreceptor cells can be classified into two distinct types, depending on whether the photopigment is borne on the membrane of a modified cilium (ciliary type) or apical microvilli (rhabdomeric type) [1]. Ciliary photoreceptors are well known as vertebrate rods and cones and are also found in several invertebrates. The rhabdomeric photoreceptor, in contrast, is a predominant type of invertebrate visual cell, but morphologically identifiable rhabdomeric photoreceptors have never been found in vertebrates. It is hypothesized that the rhabdomeric photoreceptor cell had evolved to be the photosensitive retinal ganglion cell for the vertebrate circadian photoentrainment [2, 3 and 4] owing to the fact that some molecules involved in cell differentiation are common among them [5]. We focused on the cephalochordate amphioxus because it is the closest living invertebrate to the vertebrates, and interestingly, it has rhabdomeric photoreceptor cells for putative nonvisual functions [6]. Here, we show that the amphioxus homolog of melanopsin [7, 8 and 9], the circadian photopigment in the photosensitive retinal ganglion cells of vertebrates, is expressed in the rhabdomeric photoreceptor cells of the amphioxus and that its biochemical and photochemical properties, not just its primary structure, are considerably similar to those of the visual rhodopsins in the rhabdomeric photoreceptor cells of higher invertebrates. The cephalochordate rhabdomeric photoreceptor represents an evolutionary link between the invertebrate visual photoreceptor and the vertebrate circadian photoreceptor.     

The second cytoplasmic loop of metabotropic glutamate receptor functions at the third loop position of rhodopsin.

G protein-coupled receptors identified so far are classified into at least three major families based on their amino acid sequences. For the family of receptors homologous to rhodopsin (family 1), the G protein activation mechanism has been investigated in detail, but much less for the receptors of other families. To functionally compare the G protein activation mechanism between rhodopsin and metabotropic glutamate receptor (mGluR), which belong to distinct families, we prepared a set of bovine rhodopsin mutants whose second or third cytoplasmic loop was replaced with either the second or third loop of Gi/Go- or Gq-coupled mGluR (mGluR6 or mGluR1). Among these mutants, the mutants in which the second or third loop was replaced with the corresponding loop of mGluR exhibited no G protein activation ability. In contrast, the mutant whose third loop was replaced with the second loop of Gi/Go-coupled mGluR6 efficiently activated Gi but not Gt: this activation profile is almost identical with those of the mutant rhodopsins whose third loop was replaced with those of the Gi/Go-coupled receptors in family 1 [Yamashita et al. (2000) J. Biol. Chem. 275, 34272-34279]. The mutant whose third loop was replaced with the second loop of Gq-coupled mGluR1 partially retained the Gi coupling ability of rhodopsin, which is in contrast to the fact that all the rhodopsin mutants having the third loops of Gq-coupled receptors in family 1 exhibit no detectable Gi activation. These results strongly suggest that the molecular architectures of rhodopsin and mGluR are different, although the G protein activation mechanism involving the cytoplasmic loops is common.     

Evolution of Mammalian Opn5 as a Specialized UV-absorbing Pigment by a Single Amino Acid Mutation

Opn5 is one of the recently identified opsin groups that is responsible for nonvisual photoreception in animals. We previously showed that a chicken homolog of mammalian Opn5 (Opn5m) is a G_i-coupled UV sensor having molecular properties typical of bistable pigments. Here we demonstrated that mammalian Opn5m evolved to be a more specialized photosensor by losing one of the characteristics of bistable pigments, direct binding of all-trans-retinal. We first confirmed that Opn5m proteins in zebrafish, Xenopus tropicalis, mouse, and human are also UV-sensitive pigments. Then we found that only mammalian Opn5m proteins lack the ability to directly bind all-trans-retinal. Mutational analysis showed that these characteristics were acquired by a single amino acid replacement at position 168. By comparing the expression patterns of Opn5m between mammals and chicken, we found that, like chicken Opn5m, mammalian Opn5m was localized in the ganglion cell layer and inner nuclear layer of the retina. However, the mouse and primate (common marmoset) opsins were distributed not in the posterior hypothalamus (including the region along the third ventricle) where chicken Opn5m is localized, but in the preoptic hypothalamus. Interestingly, RPE65, an essential enzyme for forming 11-cis-retinal in the visual cycle is expressed near the preoptic hypothalamus of the mouse and common marmoset brain but not near the region of the chicken brain where chicken Opn5m is expressed. Therefore, mammalian Opn5m may work exclusively as a short wavelength sensor in the brain as well as in the retina with the assistance of an 11-cis-retinal-supplying system.     

Rod Visual Pigment Optimizes Active State to Achieve Efficient G Protein Activation as Compared with Cone Visual Pigments

Most vertebrate retinas contain two types of photoreceptor cells, rods and cones, which show different photoresponses to mediate scotopic and photopic vision, respectively. These cells contain different types of visual pigments, rhodopsin and cone visual pigments, respectively, but little is known about the molecular properties of cone visual pigments under physiological conditions, making it difficult to link the molecular properties of rhodopsin and cone visual pigments with the differences in photoresponse between rods and cones. Here we prepared bovine and mouse rhodopsin (bvRh and mRh) and chicken and mouse green-sensitive cone visual pigments (cG and mG) embedded in nanodiscs and applied time-resolved fluorescence spectroscopy to compare their G_t activation efficiencies. Rhodopsin exhibited greater G_t activation efficiencies than cone visual pigments. Especially, the G_t activation efficiency of mRh was about 2.5-fold greater than that of mG at 37 °C, which is consistent with our previous electrophysiological data of knock-in mice. Although the active state (Meta-II) was in equilibrium with inactive states (Meta-I and Meta-III), quantitative determination of Meta-II in the equilibrium showed that the G_t activation efficiency per Meta-II of bvRh was also greater than those of cG and mG. These results indicated that efficient G_t activation by rhodopsin, resulting from an optimized active state of rhodopsin, is one of the causes of the high amplification efficiency of rods.     

Intramolecular Interactions That Induce Helical Rearrangement upon Rhodopsin Activation: LIGHT-INDUCED STRUCTURAL CHANGES IN METARHODOPSIN IIa PROBED BY CYSTEINE S-H STRETCHING VIBRATIONS*

Rhodopsin undergoes rearrangements of its transmembrane helices after photon absorption to transfer a light signal to the G-protein transducin. To investigate the mechanism by which rhodopsin adopts the transducin-activating conformation, the local environmental changes in the transmembrane region were probed using the cysteine S-H group, whose stretching frequency is well isolated from the other protein vibrational modes. The S-H stretching modes of cysteine residues introduced into Helix III, which contains several key residues for the helical movements, and of native cysteine residues were measured by Fourier transform infrared spectroscopy. This method was applied to metarhodopsin II_a, a precursor of the transducin-activating state in which the intramolecular interactions are likely to produce a state ready for helical movements. No environmental change was observed near the ionic lock between Arg-135 in Helix III and Glu-247 in Helix VI that maintains the inactive conformation. Rather, the cysteine residues that showed environmental changes were located around the chromophore, Ala-164, His-211, and Phe-261. These findings imply that the hydrogen bond between Helix III and Helix V involving Glu-122 and His-211 and the hydrophobic packing between Helix III and Helix VI involving Gly-121, Leu-125, Phe-261, and Trp-265 are altered before the helical rearrangement leading toward the active conformation.     

Regulatory mechanism for the stability of the meta II intermediate of pinopsin

Pinopsin is a chicken pineal photoreceptive molecule with a possible role in photoentrainment of the circadian clock. Sequence comparison among members of the rhodopsin family has suggested that pinopsin might have properties more similar to cone visual pigments than to rhodopsin, but the lifetime of the physiologically active intermediate (meta II) of pinopsin is rather similar to that of metarhodopsin II, which is far more stable than meta II intermediates of cone visual pigments [Nakamura, A. et al., (1999) Biochemistry 38, 14738-14745]. In the present study, we investigated the amino acid residue(s) contributing to this unique property of pinopsin by using site-directed mutagenesis to pinopsin-specific structural features, (i) Ser171, (ii) Asn184, and (iii) the second extracellular loop two-amino acids shorter than that of cone visual pigments. The meta II stability of the 171/184 double mutant of pinopsin (S171R/N184D) is almost the same as that of wild-type pinopsin. In contrast, the meta II lifetime is markedly shortened (one third) by introduction of the third mutation (replacement of a six-amino acid stretch, 188-193, by the corresponding eight residues of chicken green-sensitive cone pigment) to the 171/184 double mutant of pinopsin. Consistently, meta II of the green-sensitive pigment mutant, in which the eight-amino acid stretch is inversely replaced by the corresponding six residues of pinopsin, is more stable than meta II of the wild-type pigment. These results strongly suggest that the specific sequence and/or the number of residues at amino acids 188-193 in pinopsin play an important role in the stabilization of the meta II intermediate.     

Functional interaction between bovine rhodopsin and G protein transducin

To elucidate the mechanisms of specific coupling of bovine rhodopsin with the G protein transducin (G(t)), we have constructed the bovine rhodopsin mutants whose second or third cytoplasmic loop (loop 2 or 3) was replaced with the corresponding loop of the G(o)-coupled scallop rhodopsin and investigated the difference in the activation abilities for G(t), G(o), and G(i) among these mutants and wild type. We have also prepared the Galpha(i) mutants whose C-terminal 11 or 5 amino acids were replaced with those of Galpha(t), Galpha(o), and Galpha(q) to evaluate the role of the C-terminal tail of the alpha-subunit on the specific coupling of bovine rhodopsin with G(t). Replacement of loop 2 of bovine rhodopsin with that of the scallop rhodopsin caused about a 40% loss of G(t) and G(o) activation, whereas that of loop 3 enhanced the G(o) activation four times with a 60% decrease in the G(t) activation. These results indicated that loop 3 of bovine rhodopsin is one of the regions responsible for the specific coupling with G(t). Loop 3 of bovine rhodopsin discriminates the difference of the 6-amino acid sequence (region A) at a position adjacent to the C-terminal 5 amino acids of the G protein, resulting in the different activation efficiency between G(t) and G(o). In addition, the binding of region A to loop 3 of bovine rhodopsin is essential for activation of G(t) but not G(i), even though the sequence of the region A is almost identical between Galpha(t) and Galpha(i). These results suggest that the binding of loop 3 of bovine rhodopsin to region A in Galpha(t) is one of the mechanisms of specific G(t) activation by bovine rhodopsin.     

Distinct roles of the second and third cytoplasmic loops of bovine rhodopsin in G protein activation

In contrast to the extensive studies of light-induced conformational changes in rhodopsin, the cytoplasmic architecture of rhodopsin related to the G protein activation and the selective recognition of G protein subtype is still unclear. Here, we prepared a set of bovine rhodopsin mutants whose cytoplasmic loops were replaced by those of other ligand-binding receptors, and we compared their ability for G protein activation in order to obtain a clue to the roles of the second and third cytoplasmic loops of rhodopsin. The mutants bearing the third loop of four other G(o)-coupled receptors belonging to the rhodopsin superfamily showed significant G(o) activation, indicating that the third loop of rhodopsin possibly has a putative site(s) related to the interaction of G protein and that it is simply exchangeable with those of other G(o)-coupled receptors. The mutants bearing the second loop of other receptors, however, had little ability for G protein activation, suggesting that the second loop of rhodopsin contains a specific region essential for rhodopsin to be a G protein-activating form. Systematic chimeric and point mutational studies indicate that three amino acids (Glu(134), Val(138), and Cys(140)) in the N-terminal region of the second loop of rhodopsin are crucial for efficient G protein activation. These results suggest that the second and third cytoplasmic loops of bovine rhodopsin have distinct roles in G protein activation and subtype specificity.     

Molecular properties of rhodopsin and rod function

Signal transduction in rod cells begins with photon absorption by rhodopsin and leads to the generation of an electrical response. The response profile is determined by the molecular properties of the phototransduction components. To examine how the molecular properties of rhodopsin correlate with the rod-response profile, we have generated a knock-in mouse with rhodopsin replaced by its E122Q mutant, which exhibits properties different from those of wild-type (WT) rhodopsin. Knock-in mouse rods with E122Q rhodopsin exhibited a photosensitivity about 70% of WT. Correspondingly, their single-photon response had an amplitude about 80% of WT, and a rate of decline from peak about 1.3 times of WT. The overall 30% lower photosensitivity of mutant rods can be explained by a lower pigment photosensitivity (0.9) and the smaller single-photon response (0.8). The slower decline of the response, however, did not correlate with the 10-fold shorter lifetime of the meta-II state of E122Q rhodopsin. This shorter lifetime became evident in the recovery phase of rod cells only when arrestin was absent. Simulation analysis of the photoresponse profile indicated that the slower decline and the smaller amplitude of the single-photon response can both be explained by the shift in the meta-I/meta-II equilibrium of E122Q rhodopsin toward meta-I. The difference in meta-III lifetime between WT and E122Q mutant became obvious in the recovery phase of the dark current after moderate photobleaching of rod cells. Thus, the present study clearly reveals how the molecular properties of rhodopsin affect the amplitude, shape, and kinetics of the rod response.     

Photoisomerization efficiency in UV-absorbing visual pigments: protein-directed isomerization of an unprotonated retinal Schiff base

A visual pigment consists of an opsin protein and a chromophore, 11-cis-retinal, which binds to a specific lysine residue of opsin via a Schiff base linkage. The Schiff base chromophore is protonated in pigments that absorb visible light, whereas it is unprotonated in ultraviolet-absorbing visual pigments (UV pigments). To investigate whether an unprotonated Schiff base can undergo photoisomerization as efficiently as a protonated Schiff base in the opsin environment, we measured the quantum yields of the bovine rhodopsin E113Q mutant, in which the Schiff base is unprotonated at alkaline pH, and the mouse UV pigment (mouse UV). Photosensitivities of UV pigments were measured by irradiation of the pigments followed by chromophore extraction and HPLC analysis. Extinction coefficients were estimated by comparing the maximum absorbances of the original pigments and their acid-denatured states. The quantum yield of the bovine rhodopsin E113Q mutant at pH 8.2, where the Schiff base is unprotonated, was significantly lower than that of wild-type rhodopsin, whereas the mutant gave a quantum yield almost identical to that of the wild type at pH 5.5, where the Schiff base is protonated. These results suggest that Schiff base protonation plays a role in increasing quantum yield. The quantum yield of mouse UV, which has an unprotonated Schiff base chromophore, was significantly higher than that of the unprotonated form of the rhodopsin E113Q mutant, although it was still lower than the visible-absorbing pigments. These results suggest that the mouse UV pigment has a specific mechanism for the efficient photoisomerization of its unprotonated Schiff base chromophore.