3.E.1 The Ion-translocating Microbial Rhodopsin (MR) Family
Members of the MR family catalyze light-driven ion translocation across microbial cytoplasmic membranes or serve as light receptors. Among the high resolution structures for members of the MR family are the archaeal proteins, bacteriorhodopsin (Luecke et al., 1999), sensory rhodopsin II (Royant et al., 2001) and halorhodopsin (Kolbe et al., 2000) as well as an Anabaena cyanobacterial sensory rhodopsin (3.E.1.8.a) (Vogeley et al., 2004). Homologues include putative fungal chaparone proteins, a retinal-containing rhodopsin from Neurospora crassa (Maturana et al., 2001), a H+-pumping rhodopsin from Leptosphaeria maculans (Waschuk et al., 2005), retinal-containing proton pumps isolated from marine bacteria (Béjà et al., 2000), a green light-activated photoreceptor in cyanobacteria that does not pump ions and interacts with a small (14 kDa) soluble transducer protein (Jung et al., 2003; Vogeley et al., 2004) and light-gated H+ channels from the green alga, Chlamydomonas reinhardtii (Nagel et al., 2002). The N. crassa NOP-1 protein exhibits a photocycle and conserved H+ translocation residues that suggest that this putative photoreceptor is a slow H+ pump (Brown et al., 2001; see also Brown, 2004 and Waschuk et al., 2005). Allosteric structural changes in the photocycle are mediated by a sliding movement of a transmembrane helix (Takeda et al. 2004). MR proteins such as SRII exhibit fast internal motion and residual conformational entropy (O'Brien et al. 2020). Procedures for the formation of thin (mono-) and thick (multi-) layers from materials containing BR and BR/nanoparticle hybrids have been reviewed (Oleinikov et al. 2020) and their usefulness in optogenetic studies have been reviewed (Kandori 2021). The molecular determinants of ionic selectivity, photocurrent desensitization, and spectral tuning in anion- and cation-selective channelrhodopsins have been defined (Govorunova et al. 2021).
The Anabaena sensory rhodopsin exhibits light-induced interconversion between 13-cis and all trans states (Vogeley et al., 2004). The ratio of its cis and trans chromophore forms depends on the wavelength of illumination, thus providing a mechanism for a single protein to signal the color of light, for example, to regulate color-sensitive processes such as chromatic adaptation in photosynthesis. Its cytoplasmic half channel, highly hydrophobic in the archaeal rhodopsins, contains numerous hydrophilic residues networked by water molecules, providing a connection from the photoactive site to the cytoplasmic surface believed to interact with the receptor's soluble 14-kilodalton transducer.
Most proteins of the MR family are all of about the same size (250-350 amino acyl residues) and possess seven TMSs with their N-termini on the outside and their C-termini on the inside. There are 8 subfamilies in the MR family: (1) bacteriorhodopsins pump protons out of the cell; (2) halorhodopsins pump chloride (and other anions such as bromide, iodide and nitrate) into the cell; (3) sensory rhodopsins, which normally function as receptors for phototactic behavior, are capable of pumping protons out of the cell if dissociated from their transducer proteins; (4) the fungal chaparones are stress-induced proteins of ill-defined biochemical function, but this subfamily also includes a H+-pumping rhodopsin (Waschuk et al., 2005); (5) the bacterial rhodopsin, called proteorhodopsin, is a light-driven proton pump that functions as does bacteriorhodopsins; (6) the N. crassa retinal-containing receptor serves as a photoreceptor (Zhai et al., 2001); (7) the green algal light-gated proton channel, channelrhodpsin-1, (8) sensory rhodopsins from cyanobacteria and (9) light-activated rhodopsin guanylyl cyclases. A phylogenetic analysis of microbial rhodopsins and a detailed analysis of potential examples of horizontal gene transfer have been published (Sharma et al., 2006). Microbial rhodopsins have a Trp residue in the middle of TMS3, which is homologous to W86 of bacteriorhodopsin (BR), is well conserved among microbial rhodopsins with various light-driven functions, and it serves as a gate-keeper in many microbial rhodopsins (Nagasaka et al. 2020).
Bacterio- and halorhodopsins pump 1 H+ and 1 Cl- per photon absorbed, respectively. Specific transport mechanisms and pathways have been proposed (see Kolbe et al., 2000; Lanyi and Schobert, 2003; Schobert et al., 2003). The mechanism involves (1) photo-isomerization of the retinal and its initial configurational changes, (2) deprotonation of the retinal Schiff base and the coupled release of a proton to the extracellular membrane surface, and (3) the switch event that allows reprotonation of the Schiff base from the cytoplasmic side. Six structural models describe the transformations of the retinal and its interaction with water 402, Asp85, and Asp212 in atomic detail, as well as the displacements of functional residues farther from the Schiff base. The changes provide rationales for how relaxation of the distorted retinal causes movements of water and protein atoms that result in vectorial proton transfers to and from the Schiff base (Lanyi and Schobert, 2003). Helix deformation is coupled to vectorial proton transport in the photocycle of bacteriorhodopsin (Royant et al., 2000).
The marine bacterial rhodopsin has been reported to function as a proton pump. However, it most closely resembles sensory rhodopsin II of archaea as well as an Orf from the fungus Leptosphaeria maculans (AF290180). These proteins exhibit 20-30% identity with each other. Sensory rhodopsins are widespread in the microbial world, but they exhibit different modes of signaling in different organisms, including interaction with other membrane proteins, interaction with cytoplasmic transducers and light-controlled Ca2+ channel activity. Work on cyanobacteria, algae, fungi and marine proteobacteria has shown that the common design of these proteins allows rich diversity in their signaling mechanisms (Spudich 2006).
The association of sensory rhodopsins with their transducer proteins appears to determine whether they function as transporters or receptors. Association of a sensory rhodopsin receptor with its transducer occurs via the transmembrane helical domains of the two interacting proteins. There are two sensory rhodopsins in any one halophilic archaeon, one (SRI) that responds positively to orange light but negatively to blue light, the other (SRII) that responds only negatively to blue light. Each transducer is specific for its cognate receptor. An x-ray structure of SRII complexed with its transducer (HtrII) at 1.94 Å resolution is available (Gordelly et al., 2002). Molecular and evolutionary aspects of the light-signal transduction by microbial sensory receptors have been reviewed (Inoue et al. 2014).
Sol-gel immobilization of proteins in transparent inorganic matrices provide a liposomal system in which the liposome provides membrane structure. Two transmembrane proteins, bacteriorhodopsin (bR) and F0F1-ATP synthase have been incorporated into such a matrix called proteogels; if containing only bRho, a stable proton gradient forms when irradiated with visible light, whereas proteogels containing proteoliposomes with both bRho and an F0F1-ATP synthase couple the photo-induced proton gradient to the production of ATP (Luo et al. 2005). Thus, the liposome/sol-gel architecture can harness the properties of transmembrane proteins and enable a variety of applications, from power generation and energy storage to the powering of molecular motors.
Channelrhodopsin-1 (ChR1) or channelopsin-1 (Chop1; Cop3; CSOA) of C. reinhardtii is most closely related to the archaeal sensory rhodopsins. It has 712 aas with a signal peptide, followed by a short amphipathic region, and then a hydrophobic N-terminal domain with seven probable TMSs (residues 76-309) followed by a long hydrophilic C-terminal domain of about 400 residues. Part of the C-terminal hydrophilic domain is homologous to intersectin (EH and SH3 domain protein 1A) of animals (AAD30271).
Chop1 serves as a light-gated proton channel and mediates phototaxis and photophobic responses in green algae (Nagel et al., 2002). Based on this phenotype, Chop1 could be assigned to TC category #1.A, but because it belongs to a family in which well-characterized homologues catalyze active ion transport, it is assigned to the MR family. Expression of the chop1 gene, or a truncated form of this gene encoding only the hydrophobic core (residues 1-346 or 1-517) in frog oocytes in the presence of all-trans retinal produces a light-gated conductance that shows characteristics of a channel, passively but selectively permeable to protons. This channel activity may generate bioelectric currents (Nagel et al., 2002).
A homologue of ChR1 in C. reinhardtii is channelrhodopsin-2 (ChR2; Chop2; Cop4; CSOB). This protein is 57% identical, 10% similar to ChR1. It forms a cation-selective ion channel activated by light absorption. It transports both monovalent and divalent cations. It desensitizes to a small conductance in continuous light. Recovery from desensitization is accelerated by extracellular H+ and a negative membrane potential. It may be a photoreceptor for dark adapted cells (Nagel et al., 2003). A transient increase in hydration of transmembrane α-helices with a t(1/2) = 60 μs tallies with the onset of cation permeation. Aspartate 253 accepts the proton released by the Schiff base (t(1/2) = 10 μs), with the latter being reprotonated by aspartic acid 156 (t(1/2) = 2 ms). The internal proton acceptor and donor groups, corresponding to D212 and D115 in bacteriorhodopsin, are clearly different from other microbial rhodopsins, indicating that their spatial position in the protein was relocated during evolution. E90 deprotonates exclusively in the nonconductive state. The observed proton transfer reactions and the protein conformational changes relate to the gating of the cation channel (Lórenz-Fonfría et al. 2013).
Most of the MR family homologues in yeast and fungi are of about the same size and topology as the archaeal proteins (283-344 amino acyl residues; 7 putative transmembrane α-helical segments), but they are heat shock- and toxic solvent-induced proteins of unknown biochemical function. They have been suggested to function as pmf-driven chaperones that fold extracellular proteins (Zhai et al., 2001), but only indirect evidence supports this postulate. The MR family is distantly related to the 7 TMS LCT family (TC #2.A.43) (Zhai et al., 2001). It is a part of the TOG superfamily which includes G-protein coupled receptors (GPCRs) (Yee et al. 2013), and the conclusioin of homology between MRs and GPCRs has been extensively confirmed (Shalaeva et al. 2015).
Archaerhodopsin-2 (aR2), a retinal protein-carotenoid complex found in the claret membrane of Halorubrum sp. aus-2, functions as a light-driven proton pump. Trigonal and hexagonal crystals revealed that trimers are arranged on a honeycomb lattice (Yoshimura and Kouyama, 2008). In these crystals, the carotenoid bacterioruberin binds to crevices between the subunits of the trimer. Its polyene chain is inclined from the membrane normal by an angle of about 20 degrees and, on the cytoplasmic side, it is surrounded by helices AB and DE of neighbouring subunits. This peculiar binding mode suggests that bacterioruberin plays a structural role for the trimerization of aR2. When compared with the aR2 structure in another crystal form containing no bacterioruberin, the proton release channel takes a more closed conformation in the P321 or P6(3) crystal; i.e., the native conformation of protein is stabilized in the trimeric protein-bacterioruberin complex.
A crystallographic structure of xanthorhodopsin at 1.9 Å resolution revealed a dual chromophore, the geometry of the carotenoid and the retinal (Luecke et al., 2008). The close approach of the 2 polyenes at their ring ends explains why the efficiency of the excited-state energy transfer is as high as approximately 45%, and the 46 degrees angle between them suggests that the chromophore location is a compromise between optimal capture of light of all polarization angles and excited-state energy transfer. At 1.9 Å resolution, the structure revealed a light-driven proton pump with a dual chromophore. Ion-transporting rhodopsins of marine bacteria have been reviewed (Inoue et al. 2014).
Most residues participating in the trimerization are not conserved in bacteriorhodopsin, a homologous protein capable of forming a trimeric structure in the absence of bacterioruberin. Despite a large alteration in the amino acid sequence, the shape of the intratrimer hydrophobic space filled by lipids is highly conserved between aR2 and bacteriorhodopsin. Since a transmembrane helix facing this space undergoes a large conformational change during the proton pumping cycle, it is feasible that trimerization is an important strategy to capture special lipid components that are relevant to the protein activity (Yoshimura and Kouyama, 2008).
Ion-pumping bacterial rhodopsins functioning as outward H+ or Na+ and inward Cl- pumps convert light energy into transmembrane electrochemical potential differences. The H+, Na+, and Cl- pumps possess conserved respective DTE, NDQ, and NTQ motifs in helices C, which likely serve as their functional determinants, and this has been verified (Inoue et al. 2016). Phylogenetic analyses suggested that a H+ pump was the common ancestor from which Cl- pumps emerged followed by Na+ pumps. Inoue et al. 2016 proposed that successful functional conversion was achieved when these amino acid sequences changed, possibly accompanied by other changes.
Nango et al. 2016 used time-resolved serial femtosecond crystallography at an x-ray free electron laser to visualize conformational changes in bRho from nanoseconds to milliseconds following photoactivation. An initially twisted retinal chromophore displaces a conserved tryptophan residue of transmembrane helix F on the cytoplasmic side of the protein while dislodging a key water molecule on the extracellular side. The resulting cascade of structural changes throughout the protein shows how motions are choreographed as bRho transports protons uphill against a transmembrane concentration gradient.
Brho (BR) has light-independent lipid scramblase activity (Verchère et al. 2017). This activity occurs at a rate >10,000 per trimer per second, comparable to that of other scramblases including bovine rhodopsin and fungal TMEM16 proteins. BR scrambles fluorescent analogues of common phospholipids but does not transport a glycosylated diphosphate isoprenoid lipid. In silico analyses suggested that membrane-exposed polar residues in transmembrane helices 1 and 2 of BR may provide the molecular basis for lipid translocation by coordinating the polar head-groups of transiting phospholipids. Consistent with this possibility, molecular dynamics simulations of a BR trimer in a phospholipid membrane revealed water penetration along transmembrane helix 1 with the cooperation of a polar residue (Y147 in transmembrane helix 5) in the adjacent protomer. These findings suggest that the lipid translocation pathway may lie at or near the interface of the protomers of the BR trimer (Verchère et al. 2017).
Electronic current passes through bR-containing artificial lipid bilayers in solid 'electrode-bilayer-electrode' structures. The current through the protein is more than four orders of magnitude higher than would be estimated for direct tunneling through 5-nm water-free peptides. Jin et al. 2006 found that electron transport (ET) occurs only if retinal or a close analogue is present in the protein. As long as the retinal can isomerize after light absorption, there is a photo-ET effect. The contribution of light-driven proton pumping to the steady-state photocurrents is negligible. Possibly this is relevant to the early evolutionary origin of halobacteria (Jin et al. 2006).
Parvularcula oceani xenorhodopsin (PoXeR) was the first light-driven inward proton pump with a brho topology and structure, binding retinal to TMS 7. Ultrafast pump-probe spectroscopy revealed that the isomerization time of retinal is 1.2 ps, considerably slower than those of other microbial rhodopsins (180-770 fs). Following the production of J, the K intermediate was formed at 4 ps. Proton transfer occurred on a slower time-scale. While a proton was released from Asp216 into the cytoplasm, no proton-donating residue was identified on the extracellular side. A branched retinal isomerization (from 13-cis-15-anti to 13-cis-15-syn and all-trans-15-anti) occurred simultaneously with proton uptake. Thus, retinal isomerization is the rate-limiting process in proton uptake, and the regulation of pKa of the retinal Schiff base by thermal isomerization enables uptake from the extracellular medium (Inoue et al. 2018).
The generalized transport reaction for bacterio- (and some sensory) rhodopsins is:
H+ (in) + hν → H+ (out)
That for halorhodopsin is:
Cl- (out) + hν → Cl- (in)
That for xenorhodopsin is:
H+ (out) + hν → H+ (in)