Lake Baikal is the deepest and one of the most ancient lakes in the world. Its unique ecology has resulted in the colonization of a diversity of depth habitats by a unique fauna that includes a group of teleost fish of the sub-order Cottoidei. This relatively recent radiation of cottoid fishes shows a gradual blue-shift in the wavelength of the absorption maximum of their visual pigments with increasing habitat depth. Here we combine homology modeling and quantum chemical calculations with experimental in vitro measurements of rhodopsins to investigate dim-light adaptation. The calculations, which were able to reproduce the trend of observed absorption maxima in both A1 and A2 rhodopsins, reveal a Barlow-type relationship between the absorption maxima and the thermal isomerization rate suggesting a link between the observed blue-shift and a thermal noise decrease. A Nakanishi point-charge analysis of the electrostatic effects of non-conserved and conserved amino acid residues surrounding the rhodopsin chromophore identified both close and distant sites affecting simultaneously spectral tuning and visual sensitivity. We propose that natural variation at these sites modulate both the thermal noise and spectral shifting in Baikal cottoid visual pigments resulting in adaptations that enable vision in deep water light environments.
Lake Baikal is located in Eastern Siberia and it is the deepest (1600 m) lake in the world. It holds approximately one fifth of the world’s liquid freshwater. A unique feature of the lake is that oxygenation levels in even the deepest regions do not fall below 75–80% of the surface levels1. This has enabled the colonization of all depth habitats by fauna that includes a flock of teleost fish of the sub-order Cottoidei.
The Baikal cottoid fishes are an ideal system to study visual pigment evolution as both the rod and cone pigments in these fish show a gradual blue-shift in the wavelength of the absorption maxima (λmax) in relation to their habitat depth. For instance, the λmax of the rod pigment (called rhodopsin) shifts from 516 nm in the species that colonizes the surface to the 484 nm in the deepest species. These λmax shifts reflect, exclusively, variations in the amino acids interacting with the chromophore (Fig. 1a and b) as all the Lake Baikal cottoid fish utilize the same A1 chromophore (Fig. 1c)2. This is, for instance, in contrast to many freshwater teleosts where λmax red-shifts are due to the A2 chromophore3,4.
Previous studies of the Lake Baikal cottoid fish rhodopsins (from now on Baikal rhodopsins) suggest that the ancestral cottoid species that colonized the lake likely had a rhodopsin with a λmax of around 505 nm, similar to the sub-littoral species5. Variation in λmax values among present day Baikal fishes likely arose as a result of subsequent amino acid substitutions in rhodopsin, but their adaptive consequences and possible underlying mechanisms remain unclear. In deep sea fish, the observed 470–480 nm λmax is thought to be an adaptation to the blue-shifted spectral maximum of the available downwelling light6,7,8. However, these theories do not consider other aspects of rhodopsin function such as photosensitivity, which may be more important in dimly lit deepwater environments such as those found in Lake Baikal and the deep sea.
An alternative explanation for the λmax blue-shift observed in the deeper Baikal rhodopsins may be based on the existence of a Barlow-like correlation9,10. This is an inverse proportionality relationship between rhodopsin λmax and chromophore thermal isomerization rate. By competing with the photoisomerization triggering the rhodopsin function, the thermal isomerization must contribute to the thermal noise11,12 decreasing visual acuity13. When assuming the validity of such an isomerization-noise link, the Barlow correlation implies that the λmax blue-shift of the abyssal rhodopsins would reflect the need to reduce the noise in habitats with low light intensities5,14.
Here we investigate the thermal noise hypothesis through a combination of multiconfigurational quantum chemistry (MCQC) calculations and experimental studies on a set of Baikal rhodopsins. More specifically, we demonstrate the existence of a Barlow correlation between λmax, expressed in terms of the chromophore vertical excitation energy (ΔE), and isomerization rate, related to the chromophore activation energy (EaT) and derive an atomistic model of the ΔE and EaT variation. The results show that the amino acid substitutions found in the sequences of the selected Baikal rhodopsin set, modulate ΔE and EaT simultaneously in an interdependent parallel fashion suggesting that a reduction in thermal noise may have evolved in Lake Baikal fish pigments as a dim-light adaptation for increased photosensitivity.
Results and Discussion
The sequence similarity and marked λmax variation of Baikal rhodopsins facilitate the study of the effect of single amino acid substitutions on ΔE and EaT. Accordingly, we consider four species representative of different habitats (in order of depth): the littoral (1–5 m) depth (Paracottus kneri, λmax = 516 nm), the sub-littoral (1–120 m) depth (Paracottus jettelesi, λmax = 505 nm), the supra-abyssal (50–450 m) depth (Cottocomephorus inermis, λmax = 495 nm), and the abyssal (400–1500 m) depth (Abyssocottus korotneffi, λmax = 484 nm)5. For each species a MCQC-based quantum-mechanics/molecular-mechanics (QM/MM) model of the corresponding rhodopsin is constructed (Methods section and SI Appendix , Fig. S1) using, as a template,the crystallographic structure of bovine rhodopsin (Rh). The model quality is assessed by reproducing: (i) the observed λmax changes along the set plus the Rh template and (ii) the observed linear relationships between the λmax of A1/A2 pairs of pigments15,16. This second test is carried out by preparing and spectroscopically characterizing in vitro Baikal rhodopsins where the A1 chromophore is replaced by the A2 chromophore forming red-shifted analogs17. The validated QM/MM models are then used to study the effects of the amino acid substitutions differentiating the four species through a computational implementation18 of the point-charge model proposed by Nakanishi and coworkers19 (Fig. 1d and e).
Origin of the excitation energy changes
Three Lake Baikal pigments were expressed and purified in vitro with both the A1 and A2 chromophores. The measured λmax of the A1 rhodopsins were found to be almost identical to the literature values measured via microspectroscopy (MSP)2. C. inermis λmax was identical to MSP measurements (495 nm), while A. korotneffi was found to absorb maximally at 482 nm (−2 nm from MSP values) and P. jettelesi absorbed at 501 nm (−4 nm from MSP values) ( SI Appendix , Fig. S2). As expected, the λmax of the A2 rhodopsins was found to be red-shifted in comparison to the corresponding A1 rhodopsin value. A. korotneffi A2 pigment shifted to 499 nm, a total red-shift of 17 nm. P. jettelesi rhodopsin expressed with A2 chromophore shifted by 19 nm to 520 nm. C. inermis was red-shifted by 21 nm to 516 nm in the A2 pigment ( SI Appendix , Fig. S2). All A2 rhodopsins were also successfully light bleached and their MII intermediate also showed the characteristic observed blue-shifted λmax with respect to the dark adapted state ( SI Appendix , Fig. S3) as expected for functional pigments.
As reported in Fig. 2a (see also SI Appendix , Table S1), the observed A1 and A2 rhodopsin λmax trends as well as the related A1/A2 linear relationship are reproduced by the QM/MM models. Furthermore, the computed A1/A2 slope only modestly deviate from that established experimentally by Dartnall and Lythgoe15 showing a 5 nm error (i.e. <1 kcal mol−1).
In order to investigate the origin of the λmax trend, we computed the ΔE values (Fig. 2a and SI Appendix , Table S1) for the isolated (in vacuo) chromophores of the four Baikal pigments. In these computations, the geometrical parameters of the chromophore are fixed at the values of the S0 equilibrium structure of the QM/MM model. The results provide information on the ΔE variations due to the changes in chromophore geometry. Within the A1 and A2 sets, the ΔE values show only limited ≤1 kcal mol−1 variations consistently with the limited geometrical changes displayed in Fig. 2b (i.e. with dihedral angle changes ≤4 degrees). Thus, the model indicates that the λmax variations are not due to progressive chromophore distortion (except for a fraction in the case of C. inermis) and must be dominated by electrostatic effects (i.e. by the variations in the point charges of cavity and extra-cavity amino acids).
Effect of cavity and extra-cavity amino acids
The ΔE change between the most red-shifted model (P. kneri) and the most blue-shifted model (A. korotneffi) is 1.9 and 3.3 kcal mol−1 for the A1 and A2 chromophore respectively. This value (see Fig. 1d) reflects the stabilizing effect of the P. kneri and A. korotneffi protein environments on the difference in S1 and S0 charge distribution of the chromophore (the S1/S0 charge difference of Fig. 2c). Since the S1/S0 charge difference is similar in all pigments, we focused on the larger ΔE changes of the A2 rhodopsins.
The ΔE decrease (red-shift) or increase (blue-shift) associated with a specific side-chain, can be evaluated by setting its point charges to zero and recomputing the excitation energy (ΔEoff). The largest ΔE-ΔEoff differences computed for the cavity residues are displayed in the balloon diagrams of Fig. 3a. When comparing the effects of side-chain substitutions, one finds that a ΔE change may have two components. The first is a direct component due to the change in number, magnitude and position of the corresponding side-chain point charges. The second component is indirect and originates from the reorganization of the hydrogen bond network (HBN) induced by the same substitution. This second component/effect explains why conserved residues and water molecules may display large ΔE-ΔEoff changes and contribute to the total ΔE variation significantly.
When comparing the extreme cases of P. kneri (reddest) and A. korotneffi (bluest), the sequence data shows that the amino acid substitutions G114A and Y261F remove two red-shifting residues in P. kneri (see Fig. 3a) which directly contribute to blue-shifting the A. korotneffi absorption. While the same data shows that A292S does not change the ΔE-ΔEoff, below we will see that this substitution modifies the HBN which then blue-shifts the λmax indirectly. Thus variations in the composition of the rhodopsin cavity modulates the λmax between littoral and abyssal habitats through direct and indirect changes. The same analysis indicates that, due to a cancellation of ΔE-ΔEoff of opposite signs (e.g. the sizable R140C red-shifting replacement is counterbalanced by the smaller T209I, L176S, T297S blue-shifting replacements in Fig. S6), the substitution of extra-cavity residues contributes only modestly to the λmax change from P. kneri to A. korotneffi.
The sub-littoral and supra-abyssal species P. jettelesi and C. inermis feature the same amino acid cavity composition and similar cavity ΔE-ΔEoff values ( SI Appendix , Fig. S6). In contrast, the extra-cavity substitutions T297S, D83N, T166S relating these species are associated with direct blue-shifting changes. This suggests that spectral tuning among species in the closer sub-littoral and supra-abyssal habitats may be controlled by extra-cavity amino acids. On the other hand, the ΔE variations computed between sub-littoral and littoral and between abyssal and supra-abyssal are modulated by both cavity and extra-cavity substitutions and by direct and indirect changes ( SI Appendix , Table S4 and S5) as we will discuss below.
Activation energy changes
In order to find out if the blue-shift observed when passing from the littoral to the abyssal habitat reflects the need to reduce the rhodopsin thermal noise, we built the QM/MM models for the S0 transition states (TS, Fig. 1e) that control thermal isomerization. The models allow to compute the corresponding EaT, thermal activation energy. The results yield a linear relationship between EaT and 1/λmax (Fig. 2d) with the most blue-shifted A1 rhodopsin (from A. korotneffi) displaying an EaT 5.4 kcal mol−1 higher than the EaT of the most red-shifted rhodopsin (from P. kneri). Notice that the present work is not aimed at reproducing the absolute values of the observed barriers but only their variation among different Baikal species. This is discussed in Section 6 of the SI Appendix which highlights a non-Arrhenius behavior as a source of discrepancy between computed and available observed EaT values. In the same section, an additional source of inaccuracy is associated with the fact that reactant and transition state structures are computed as single points on the rhodopsin potential energy surface without explicitly accounting for the protein dynamics at body temperature. However, this error is expected to be systematic and therefore unable to affect the computed trends.
The geometrical structures of the chromophore at the TSs of A. korotneffi and P. kneri (see Fig. 2e) at the transition state are similar and consistent with those reported for Rh18. The structures indicate that the isomerization occurs via an aborted bicycle-pedal reaction coordinate20,21 involving the -C9=C10-C11=C12- segment of the chromophore backbone. Such motion is coupled with a substantially complete charge translocation from the =C12-C13=C14-C15=NH- segment to the segment containing the β-ionone ring (compare the schematic S0 reactant - i.e. the dark adapted state - and TS structures in Fig. 1d and e respectively).
Similar to what was found for ΔE, the EaT of the chromophores in vacuo, i.e. the energy difference between the chromophores extracted from the QM/MM models of the TS and S0 reactant, are close (see Fig. 2d and Table S6). It is therefore concluded that the changes in EaT are due to variations in the protein environment. Furthermore in Fig. 2d we show that electrostatic interactions prevail over steric (e.g. van der Waals) interactions. In order to isolate the steric effects, we zeroed all protein charges of the models and recomputed the EaT values. The A. korotneffi value is found to be lower than the corresponding P. kneri value showing that steric effects would, as confirmed by the P. jettelesi and C. inermis EaT values, result in a trend opposite to the one observed when both steric and electrostatic effects are considered (see also SI Appendix ). It is thus concluded that the protein electrostatics determines the EaT trend.
According to the point charge model, EaT is modulated by the residue charges which “stabilize” or “destabilize” the TS/S0 charge changes (Fig. 2f). Such difference is qualitatively similar to the S1/S0 charge change (compare Fig. 2c and f). Thus, we investigate the differences in EaT between P. kneri and A. korotneffi by applying the same analysis employed for ΔE. Accordingly, the effect of each residue is evaluated by computing the quantity EaT-EaToff (EaToff being the barrier obtained after zeroing the charges of a specific residue). When a residue is replaced such quantity is expected to display variations similar to the one seen for ΔE-ΔEoff but the significance of which is more complex to interpret. In fact, while ΔE-ΔEoff reflects, by definition, the effect of the residue charges, EaT-EaToff also incorporates the effect of the geometrical difference between the TS and the S0 reactant. The EaT-EaToff variations induced by the rhodopsin cavity substitutions relating P. kneri to A. korotneffi (Y261F, A292S and G114A), are given in Fig. 3b and c and Table S7. Y261F leads, through a direct change, to an increase of EaT in A. korotneffi (2.7 kcal mol−1) consistently with the effect reported above for ΔE. As shown in Fig. 3c A292S leads, again through a direct change, to a large increase (4.8 kcal mol−1) in EaT of A. korotneffi. Although this variation parallels the corresponding ΔE increase, the modeled EaT change is due to HBN modification rather than a direct change as for ΔE. Finally, while G114A (see Fig. S7) leads to a negligible EaT variation in A. korotneffi, it causes a limited direct ΔE increase (0.8 kcal mol−1). In conclusion, while the overall variation induced by the three substitutions show the same trend for both ΔE and EaT, their contributions may be mechanistically distinct as we detail below.
Mechanisms of thermal noise modulation and spectral tuning
As reported above the combined effects of three cavity substitutions (see Fig. 4a) play a substantial role in establishing the differences between the ΔE and EaT of P. kneri and A. korotneffi. The Y261F substitution blue-shifts the λmax of all species relative to P. kneri by effectively changing the side-chain point charges. In fact, Y261F loses a dipole (the OH group of tyrosine) pointing its negative pole towards the β-ionone ring (see Fig. 4a top). This destabilizes the S1/S0 charge difference of Fig. 2c increasing the ΔE and leading to a blue-shift. As shown in Fig. 4b the same mechanism is seen when comparing P. kneri and P. jettelesi.
A parallel mechanism explains the increase of EaT in A. korotneffi. with respect to P. kneri. In fact, similar to the ΔE effect, the Y261F substitution in A. korotneffi destabilizes the TS/S0 charge shift laid out in Fig. 2f and thus increases the EaT. However, in contrast to ΔE, EaT is also modulated via an indirect effect of the same Y261F substitution. In fact, the loss of OH in position 261 in A. korotneffi, which used to form a hydrogen bond with the backbone oxygen of the conserved G121 residue in P. kneri (compare bottom and top in Fig. 3b), induces an HBN change. This change affects the stability of the TS and S0 reactant differently and contributes to the EaT increase in A. korotneffi.
As seen in Fig. 3c, the A292S substitution relating P. kneri to A. korotneffi does not blue-shift the λmax through a direct change, but through a modification of the HBN. In fact, A292S induces a relocation/reorientation of WAT2 which displaces it away from the Schiff base region (see Figs 3c and 4a bottom). Since the positive pole of WAT2 points towards the -C15=NH- moiety and destabilizes the S1/S0 charge difference, such WAT2 relocation increases the ΔE in A. korotneffi. The same mechanism, which is also responsible for the P. jettelesi to A. korotneffi λmax blue-shift (see Fig. 4c), explains the increased EaT in A. korotneffi through a decreased destabilization of the TS/S0 charge difference. However, the A292S induced WAT2 relocation also mediates a secondary indirect change of EaT. In fact, it perturbs an HBN connecting the conserved residues E181, S186 and Y268 (see bottom and top in Fig. 4b) which thus contribute to modulate EaT. This is demonstrated by the 1.3 kcal mol−1 increase of S186 and −2.1 and −1.4 kcal mol−1 decrease of E181 and Y268 respectively in A. korotneffi compared to P. kneri. Notice that, although individually E181 and Y268 induce a reduction in EaT, such HBN modulation is dominated by the 4.6 kcal mol−1 increase due to WAT2 (see Fig. 3c).
Finally, the G114A substitution, which replaces a non-polar residue with a sterically larger residue, shows a contrasting effect in A. korotneffi. As shown in Fig. 4a top and c, the G114 hydrogen of P. kneri and P. jettelesi is close to the Schiff base linkage and stabilizes the S1/S0 charge difference. Thus, the G114A substitution in A. korotneffi contributes to increase the ΔE. Such ΔE change is not paralleled EaT which instead decreases. Nevertheless, due to the limited change in polarity, the decrease (see Fig. S7) is smaller than the EaT increase due to the Y261F and A292S substitutions.
In conclusion, point-charge analysis has revealed a set of substitutions which simultaneously modulate ΔE and EaT via cooperative direct and indirect HBN mediated mechanisms. While the magnitude of the described changes is expected to be sensitive to the details of our basic QM/MM models, the same substitutions have been detected in other contexts. In fact, Y261F has been shown to be responsible for the spectral differentiation between green and red cone pigments in primates22. G114A has been shown to cause a blue-shift also in Rh when expressed in vivo23,24 and in spite of the limited polarity change. A292S has also been detected in blue-shifted rhodopsin from other fish25, marine mammals26, and monotremes27.
Light-sensitivity in related species
Above we have employed MCQC-based QM/MM models of rhodopsins reconstituted with both A1 and A2 retinals to investigate the relationship between spectral tuning and thermal isomerization rate in different species of cottoid fish. The results support the existence of a direct proportionality relationship between ΔE and EaT for pigments of closely related species which evolved in the confined environment of Lake Baikal. This expands the validity of the Barlow correlation discussed for rod and cone pigments of distant species9,10,28 and provides a link with the observed inverse proportionality relationship between λmax and isomerization rate in proton-pumping rhodopsins29 and even in the extreme case of 13-cis retinal chromophore salts in solution30.
The ΔE and EaT proportionality originates at the electronic level. Indeed, the similarity between the S1/S0 and TS/S0 charge differences, (see Fig. 2c and f) due to the changes in chromophore π-electron density, makes ΔE and EaT sensitive to the same substitutions. At a more fundamental level, such similarity originates from the fact that the same charge transfer configuration (ϕCT) of the chromophore dominates the rhodopsin vertical S1 state and S0 transition state. As previously shown18, this is a consequence of a quantum mechanical property of the conical intersection of the rhodopsin chromophore18,31. Therefore the λmax changes observed in Baikal rhodopsins reflects the biological exploitation of a quantum effect to increase light sensitivity32.
The analysis of the QM/MM models indicates that the variation of ΔE and EaT in phylogenetically closely related rhodopsins is controlled by the electrostatic characteristics of the protein. Our implementation of Nakanishi’s point charge analysis has identified 8 rhodopsin substitutions, over a total of 20, modulating light sensitivity from red-shifted P. kneri to the blue-shifted A. korotneffi. The same analysis also produced an “atomistic model” of dim-light adaptation through specific side-chain substitutions. Through this model, specific mechanisms can be associated to the proposed phylogeny5 assumed to originate from P. jettelesi as its λmax matches that of the ancestor. While the modification of the point charges associated with a cavity substitution have a direct impact on ΔE and EaT (e.g. F261Y when comparing P. jettelesi and P. kneri in Fig. 4b), it would be impossible to model the observed trends without taking into account the HBN modifications associated with the same substitution (e.g. A292S comparing P. jettelesi and P. korotneffi in Fig. 4c) or the effect of extra-cavity substitutions (e.g. D83N, T297S and T166S when comparing P. jettelesi and C. inermis and, additionally, S298A in P. jettelesi and A. korotneffi). Also, in our QM/MM models, extra-cavity substitutions display large effects when an ionized residue replaces a neutral one (e.g. C140R replacing cysteine in P. jettelesi to a arginine in P. kneri).
In conclusion, when assuming that the thermal isomerization of rhodopsin dominates its thermal noise, the regular Baikal rhodopsin blue-shift observed when moving from littoral to abyssal habitats may be a byproduct of visual adaptations to extremely low levels of illumination. In fact, our study shows that for Baikal fishes, these two aspects of visual pigment function are interdependent: the isomerization rate (which would determine the amount of thermal noise) and the wavelength of maximal absorbance. Amino acid substitutions have evolved in these fishes that shift both quantities simultaneously for adaptations that would contribute to better visual sensitivity and enable colonization of the dimly lit blue-shifted deepwater environments of Lake Baikal. Our results suggest that it is possible similar mechanisms may underlie colonization of other deepwater dimly lit environments such as those inhabited by deep sea fishes in marine habitats.
Molecular biology methods
No experiments on live vertebrates were carried out in this study. Incomplete Baikal cottoid RH1 sequences were taken from5 and completed with wildtype bovine sequences for the N- and C-termini. The full-length hybrid RH1 genes were synthesized by GeneArt (Invitrogen) with BamHI and EcoRI restriction sites at the 5′ and 3′ ends, respectively. The synthesized sequences were then inserted into the p1D4-hrGFP II expression vector which tags expressed rhodopsin sequences with the nine amino acid 1D4 peptide sequence (TETSQVAPA) at the carboxy terminus33. This enables immunoaffinity purification of expressed proteins from HEK293T cells as previously described34,35. UV-vis absorption spectra of purified rhodopsin samples were measured at room temperature both in the dark, and following light-bleaching for 60 seconds using a fiber optic lamp. Difference spectra were calculated by subtracting the light-bleached spectra from respective dark spectra. To provide accurate estimates of λmax, dark absorbance spectra were fit to standard templates for either A1 or A2 visual pigments36.
The QM/MM models of both A1 and A2 fish rhodopsins were prepared starting with a structures obtained via comparative modeling. To do so, the chain A of the 1U19 structure of bovine rhodopsin37 was used as a template. The models were then constructed by relaxing the cavity-counterion-chromophore complex in its protein environment via molecular dynamics and geometry optimization. The chromophore was treated using the complete-active-space self-consistent field (CASSCF) method38 with an active space corresponding to the entire π-system and the 6–31G* basis set. The protein environment was instead described using the AMBER force field. To account for the dynamic electron correlation, the model equilibrium CASSCF/AMBER geometries and wavefunctions were used for single-point multiconfigurational second-order perturbation theory (CASPT2) calculations with a two-root state average zeroth-order wavefunction39. The ΔE values are computed at the CASPT2//CASSCF/AMBER level. The transition states controlling the thermal isomerization were located via restricted-step rational-function-optimizations40 at the CASSCF/AMBER level. The corresponding EaT values were computed at the CASPT2//CASSCF/AMBER level. See the SI Appendix for further details.
How to cite this article: Luk, H. L. et al. Modulation of thermal noise and spectral sensitivity in Lake Baikal cottoid fish rhodopsins. Sci. Rep. 6, 38425; doi: 10.1038/srep38425 (2016).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Dundas, J. et al. CASTp: computed atlas of surface topography of proteins with structural and topographical mapping of functionally annotated residues. Nucleic. Acids. Res. 34, W116–W118 (2006).
Bowmaker, J. K. et al. Visual pigments and the photic environment: the cottoid fish of Lake Baikal. Vision. Res. 34, 591–605 (1994).
Crescitelli, F. The visual cells and visual pigments of the vertebrate eye Photochemistry of vision (ed. Dartnall, H. J. A. ) 245–363 (Springer, 1972).
Bridges C. D. B. The rhodopsin-porphyropsin visual system. Photochemistry of Vision (ed. Dartnall, H. J. A. ) 417–480 (Springer, 1972).
Hunt, D. M., Fitzgibbon, J., Slobodyanyuk, S. J. & Bowmakers, J. K. Spectral tuning and molecular evolution of rod visual pigments in the species flock of cottoid fish in Lake Baikal. Vision. Res. 36, 1217–1224 (1996).
Clarke, G. L. On the depth at which fish can see. Ecology. 17, 452–456 (1936).
Crescitelli, F., McFall-Ngai, M. & Horwitz, J. The visual pigment sensitivity hypothesis: further evidence from fishes of varying habitats. J. Comp. Physiol. A. 157, 323–333 (1985).
Partridge, J. C., Shand, J., Archer, S. N., Lythgoe, J. N. & van Groningen-Luyben, W. A. Interspecific variation in the visual pigments of deep-sea fishes. J. Comp. Physiol. A. 164, 513–529 (1989).
Barlow, H. B. Purkinje shift and retinal noise. Nature 179, 255–256 (1957).
Ala-Laurila, P., Pahlberg, J., Koskelainen, A. & Donner, K. On the relation between the photoactivation energy and the absorbance spectrum of visual pigments. Vision. Res. 44, 2153–2158 (2004).
Baylor, D. A., Matthews, G. & Yau, K. W. Two components of electrical dark noise in toad retinal rod outer segments. J. Physiol. 309, 591–621 (1980).
Yanagawa, M. et al. Origin of the low thermal isomerization rate of rhodopsin chromophore. Sci. Rep. 5, 11081 (2015).
Aho, A.-C., Donner, K., Hyden, C., Larsen, L. O. & Reuter, T. Low retinal noise in animals with low body temperature allows high visual sensitivity. Nature 334, 348–350 (1988).
Donner, K., Firsov, M. L. & Govardovskii, V. I. The frequency of isomerization-like ‘dark’ events in rhodopsin and porphyropsin rods of the bull-frog retina. J. Physiol. 428, 673–692 (1990).
Dartnall, H. J. A. & Lythgoe, J. N. The spectral clustering of visual pigments. Vision. Res. 5, 81–100 (1965).
Hárosi, F. I. An analysis of two spectral properties of vertebrate visual pigments. Vision. Res. 34, 1359–1367 (1994).
Wald, G., Brown, P. K. & Smith, P. H. Cyanopsin, a new pigment of cone vision. Science 118, 505–508 (1953).
Gozem, S., Schapiro, I., Ferré, N. & Olivucci, M. The molecular mechanism of thermal noise in rod photoreceptors. Science 337, 1225–1228 (2012).
Nakanishi, K., Balogh-Nair, V., Arnaboldi, M., Tsujimoto, K. & Honig, B. An external point-charge model for bacteriorhodopsin to account for its purple color. J. Am. Chem. Soc. 102, 7945–7947 (1980).
Frutos, L. M., Andruniów, T., Santoro, F., Ferré, N. & Olivucci, M. Tracking the excited-state time evolution of the visual pigment with multiconfigurational quantum chemistry. Proc. Natl. Acad. Sci. USA 104, 7764–7769 (2007).
Schapiro, I. et al. The ultrafast photoisomerizations of rhodopsin and bathorhodopsin are modulated by bond length alternation and HOOP driven electronic effects. J. Am. Chem. Soc. 133, 3354–3364 (2011).
Neitz, M., Neitz, J. & Jacobs, G. H. Spectral tuning of pigments underlying red-green color vision. Science 252, 971–974 (1991).
Han, M., Lin, S. W., Smith, S. O. & Sakmar, T. P. The effects of amino acid replacements of glycine 121 on transmembrane helix 3 of rhodopsin. J. Biol. Chem. 271, 32330–32336 (1996).
Rader, A. J. et al. Identification of core amino acids stabilizing rhodopsin. Proc. Natl. Acad. Sci. USA 101, 7246–7251 (2004).
Sugawara, T. et al. Parallelism of amino acid changes at the RH1 affecting spectral sensitivity among deep-water cichlids from Lakes Tanganyika and Malawi. Proc. Natl. Acad. Sci. USA 102, 5448–5453 (2005).
Fasick, J. I. & Robinson, P. R. Spectral-tuning mechanisms of marine mammal rhodopsins and correlations with foraging depth. Vis. Neurosci. 17, 781–788 (2000).
Bickelmann, C., Morrow, J. M., Müller, J. & Chang, B. S. Functional characterization of the rod visual pigment of the echidna (Tachyglossus aculeatus), a basal mammal. Vis. Neurosci. 29, 211–217 (2012).
Luo, D. G., Yue, W. W., Ala-Laurila, P. & Yau, K. W. Activation of visual pigments by light and heat. Science 332, 1307–1312 (2011).
Milder, S. J. Correlation between absorption maxima and thermal isomerization rates in bacteriorhodopsin. Biophys. J. 60, 440–446 (1991).
Sheves, M. & Baasov, T. Factors affecting the rate of thermal isomerization of 13-cis-bacteriorhodopsin to all trans. J. Am. Chem. Soc. 106, 6840–6841 (1984).
Herzberg, G. & Longuet-Higgins, H. C. Intersection of potential energy surfaces in polyatomic molecules. Discuss. Faraday Soc. 35, 77–82 (1963).
Lambert, N. et al. Quantum biology. Nat. Phys. 9, 10–18 (2013).
Morrow, J. M. & Chang, B. S. The p1D4-hrGFP II expression vector: a tool for expressing and purifying visual pigments and other G protein-coupled receptors. Plasmid 64, 162–169 (2010).
Morrow, J. M., Lazic, S. & Chang, B. S. A novel rhodopsin-like gene expressed in zebrafish retina. Vis. Neurosci. 28, 325–335 (2011).
Morrow, J. M. & Chang, B. S. Comparative Mutagenesis Studies of Retinal Release in Light-Activated Zebrafish Rhodopsin Using Fluorescence Spectroscopy. Biochemistry 54, 4507–4518 (2015).
Govardovskii, V. I., Fyhrquist, N., Reuter, T., Kuzmin, D. G. & Donner, K. In search of the visual pigment template. Vis. Neurosci. 17, 509–528 (2000).
Okada, T. et al. The retinal conformation and its environment in rhodopsin in light of a new 2.2 A crystal structure. J. Mol. Biol. 342, 571–583 (2004).
Roos, B. O. et al. Ab Initio Methods in Quantum Chemistry, II Advances in Chemical Physics (ed. Lawley, K. P. ) 399–446 (John Wiley & Sons: Chichester, UK, 1987).
Andersson, K., Malmqvist, P. A., Roos, B. O., Sadlej, A. J. & Wolinski, K. Second-order perturbation theory with a CASSCF reference function. J. Phys. Chem. 94, 5483–5488 (1990).
Besalú, E. & Bofill, J. M. On the automatic restricted-step rational-function-optimization method. Theor. Chem. Acc. 100, 265–274 (1998).
The authors are indebted to Victor I Govardovskii for many helpful comments. This work was supported in part by the National Science Foundation under grants no. CHE-1152070 and CHE-1551416 and the Human Frontier Science Program Organization under grant RGP0049/2012. M.O. is grateful to the Center for Photochemical Sciences and School of Arts and Sciences of the Bowling Green State University. The authors are indebted to NSF-XSEDE and Ohio Supercomputer Center for granted computer time.
The authors declare no competing financial interests.
Electronic supplementary material
About this article
Cite this article
Luk, H., Bhattacharyya, N., Montisci, F. et al. Modulation of thermal noise and spectral sensitivity in Lake Baikal cottoid fish rhodopsins. Sci Rep 6, 38425 (2016). https://doi.org/10.1038/srep38425
Nature Communications (2020)
Nature Chemistry (2018)