Retinal orientation and interactions in rhodopsin reveal a two-stage trigger mechanism for activation

The 11-cis retinal chromophore is tightly packed within the interior of the visual receptor rhodopsin and isomerizes to the all-trans configuration following absorption of light. The mechanism by which this isomerization event drives the outward rotation of transmembrane helix H6, a hallmark of activated G protein-coupled receptors, is not well established. To address this question, we use solid-state NMR and FTIR spectroscopy to define the orientation and interactions of the retinal chromophore in the active metarhodopsin II intermediate. Here we show that isomerization of the 11-cis retinal chromophore generates strong steric interactions between its β-ionone ring and transmembrane helices H5 and H6, while deprotonation of its protonated Schiff's base triggers the rearrangement of the hydrogen-bonding network involving residues on H6 and within the second extracellular loop. We integrate these observations with previous structural and functional studies to propose a two-stage mechanism for rhodopsin activation.

While the experimental results are very solid and interesting, and must be published, this reviewer believes that the integrated hypothesis of a two-stage rhodopsin activation is not presented convincingly. While the proposed mechanism may be 100% correct, the presentation of this mechanism, and the way it integrates the experimental data with the vast body of the previous knowledge, need to be improved. In the present form, it is inaccessible to anyone not in the immediate narrow field, the abstract and the discussion look disconnected from the experimental results, and the whole paper looks somewhat disjointed. More specifically, the following points will be unclear for most readers: i) why the retinal proximity data are collected on Meta II only (not on Meta I), but used to argue for the stage I of the triggering; ii) why and how the NMR data collected just for a few residues are used to build much more general mechanism. In other words, the experimental results should be put into the context of what is already known much more explicitly, and the new and old results should be contrasted better; iii) the FTIR data are not integrated well with the NMR data. This piece looks foreign, unless the integration is explained better. Additionally, better description of what was actually done is needed. Not a single FTIR spectrum is shown. iv) Methods description is very incomplete. For example, the conditions for Meta I are not given anywhere, chemical shifts for slices are often missing, FTIR reference spectra are not there, etc.; v) Similarly, many statements (which may look obvious to the authors, but not to most readers) are not supported by references. More specific examples of these points are listed below. In my opinion, the paper would be much stronger if the focus were shifted from the "high level integration" two stage hypothesis to the discussion of the true orientation of retinal in Meta II in the context of associated protein changes. On the other hand, if the focus remains the same, the hypothesis should be much better integrated with the presented results and placed in context of the existing knowledge.
Specific points, some of which are just minor editorial issues and some illustrate the general points mentioned above: 1) p. 1, address #1 is incomplete 2) p. 2. abstract -see above. The main point that the proposed hypothesis integrates a lot of the old data with some of the new data presented in the paper is missing. A better placement into the context would be helpful. 3) p. 3, "As a result, it has been a challenge" -the logic is not clear. 4) p. 5, "The NMR and FTIR approaches make use of low temperature to trap the active Meta-II state" -to be fair, one can argue that NMR experiments are done on the frozen detergent micelles, which is not completely native. On the other hand, it seems that FTIR results were obtained on the HEK cell membranes (even though the methodology description is not very clear here). The difference in sample conditions between NMR and FTIR is not discussed. 5) p. 7, "the relative intensity of the cross peak to the Phe261" -relative to what? 6) p. 7, reference to Fig. S2, actually refers to Fig. S2B. The data presented in Fig. S2A are never discussed or referred to in the main text. 7) p. 7, "lost, consistent with an increased separation between these tyrosines and the retinal C19 carbon" -an alternative explanation would be retinal rotation, which should be mentioned and argued against once more. 8) p. 7, "loss of intensity of these tyrosines with the retinal C19 methyl resonance" -style 9) p. 8, "The row through the C20 diagonal resonance yields" -here and elsewhere, it should be clearly stated that this resonance may change between Rd and Meta II and the actual positions indicated in the figures. 10) p. 9, "Tyr2686.51 has the highest subfamily conservation" -elsewhere in the paper it says "one of the highest", also what about W265? 11) p. 11, "This cross peak does not lose intensity as previously assigned" -style 12) p. 11, the discussion of Fig. 2 peak intensity changes needs to be more quantitative. E.g., what is the intensity increase for the 268/122 crosspeak? 13) p. 11, "Both increase to > 6 Å" -style 14) p. 12, "Tyr2686.51 on TM helix H6 is strongly hydrogen bonded to Glu181" -indicate according to which structure 15) p. 12, "and that Tyr191EL2 has shifted relative to Glu181EL2" -not clear where this came from, as E181 has not been observed in this paper. Needs better explanation. 16) p. 13, "against Ala2726.55 H6" -style 17) p. 13, "the inward tilt of H6" -confusing, as there is also outward tilt and rotation of H6 mentioned elsewhere. Would be nice to specify which half of the helix is involved for clarity. 18) p. 13, it would be nice to have a schematics of the proposed changes in the EL2 cluster 19) p. 14, "The observation that Tyr191EL2 and Tyr2686.51 strongly stabilize the Meta-I state suggests that in crystal structures of opsin and/or Meta-II, the inactive hydrogen-bonding network reforms in this region of the receptor upon the decay of Meta-II to opsin" -needs better explanation 20) p. 15, the first subsection of the discussion is not needed as a separate section, in my view, it mainly repeats introduction. 21) p. 15, "mechanism is highlighted by the ~35 kcal/mol" -not clear 22) p. 15, "Our current results show this change is mediated by interactions of the retinal with Phe2616.44 and Trp2656.48" -not clear as W265 has not been detected in this work. 23) p. 16, discussion on the interaction of His211 and Glu122 -once again, it is not clear where this comes from. Glu122 has not been measured in this work, if this is from the old data, the references should be given. 24) p. 16, "results in motion of the retinal PSB proton away from the stabilizing interaction with its counter-ion Glu113" -a bit unorthodox, as normally people speak about the PSB counterion rather than proton counterion. 25) p. 17, "Upon isomerization, the C19 methyl group rotates in the opposite direction" -reference is needed 26) p. 19, "13C label rhodopsin" -style 27) Fig. 1 legend -PDB accession should be given; positions of the diagonal resonances used should be indicated; "scaled to the C12-C20 cross peaks" -specify how; 28) Fig. 1H -Signal-to-noise is poor resulting in some negative bands (phasing?). May be comment on that? 29) Fig. 3 legend -"are shown of rhodopsin minus Meta-I (orange) and rhodopsin minus Meta-II (black) of rhodopsin" -style 30) Fig. 3 -large perturbations of Met and Tyr other than those discussed are observed, but ignored. Assignments of those resonances are not explained (reference if from the previous work) 31) Fig. 4 legend refers to methods for the reference FTIR spectra, but those are not described there. 32) Fig. 3 -two panels are labeled C 33) Suppl. p. 4 -"the 3Cε-Met" 34) Suppl. p. 5 -for the discussion of the ring inversion and other possible changes in retinal (resulting in C18-Phe261 proximity) a figure would be nice 35) Suppl. Fig. S4 legend and the following paragraph overlap significantly. May be merge the two. 36) Suppl. p. 9 -two references to non-existing Fig. 3E are made. 37) Suppl. p. 12 -"absence of a cross peak between the retinal chromophore and 13Cζ-labeled tyrosine in Meta-II" -not true in general, should specifically name the retinal part discussed 38) Suppl. Fig. S6 -normalization of the C5-C18 cross peak is confusing, as its amplitude decreases 39) Suppl. p. 13 -(S.O. Smith, personal communication) -not sure one can refer to personal communication from one of the authors. May be unpublished is more appropriate. 40) Suppl. p. 13 -"The high frequency of the 13C5 chemical shift suggests that the retinal with the flipped ring orientation is bound to Lys296 as a protonated Schiff's base" -explain and reference 41) Suppl. p. 14 -I would move some of the very interesting discussion on Meta III into the main text. 42) Suppl. Fig. 7 -the legend needs tightening, it is not very clear and accessible.
Reviewer #2 (Remarks to the Author): This manuscript reports the results of a very informative, if not groundbreaking, study on vertebrate rhodopsin. SS-NMR dipolar assisted rotation resonance (DARR) along with 13C isotope labeling of the retinal at specific positions and/or phenylalanine and tyrosine was used to measure distances between specific atoms in the protein. Although this method cannot pick up cross peaks when distances are more than 6-6.5 Å, , absence of such peaks along with signal strength when detected can be used to infer the proximity of specific groups, especially when comparing the structure of the dark state of rhodopsin with bleaching intermediates.
One major conclusion concerns the orientation of the β-ionone ring in the dark-state and Meta-II intermediate. While the activated state structure has been elucidated for opsinm it suffers from the lack of a retinal chromophore thus precluding direct visualization of how the retinal isomerization triggers the conversion to the active state. In two structures of Meta-II obtained using crystals of the M257Y mutant and opsin containing soaked-in retinal, the orientation of the ring has flipped from the dark-state. However, the DARR experiment, which measures the low-temperature trapped active state of rhodopsin photointermediate shows convincingly that the β-ionone ring does not flip under these conditions. The paper also provides detailed insight into the molecular events which lead to the transition to Meta II including outward rotation of helix-6, a key feature of GPCRs, and conveyance of a signal to EL2. This includes new information on the interaction of Tyr268 on helix-6 with Tyr191 (β-4 strand on EL2) and Glu181 (β-3 strand of EL2 which in the dark state interacts with the SB). An FTIR pH difference technique developed by the Vogel group to study the transition from Meta-I to Meta-II (and sub-states) also provides additional support for some of the conclusions.
Overall, this is an innovative, technically sound study which provides important new insights into structural changes leading to rhodopsin activation, the paper will be of wide interest and will stimulate studies of other ligand-activated GPCRs.
There are a number of minor problems with the manuscript which the authors need to correct before publication:  Figure 1B. Note that Tyr191 does not appear to be the closest residue to retinal C18. If Figure 1B is the dark state then this is understandable but in that case one of the Meta II forms of the crystal structure should also be shown.
Page 6: It should be noted that the referred to peak at 21.6 ppm is not shown. Figure 1C: In bottom row of DARR data add lines to show the Phe261-C18 cross peaks as done for DARR row above. Page 7, paragraph 2: Thr118 is not shown in Figure 1B     The authors present an extensive solid-state NMR study supported by FTIR and mutagenesis on rhodopsin. The aim of their work is to show that the activation of rhodopsin requires multiple steps of interaction between the retinal chromophore and opsin after the isomerisation event. Recent Xray studies have shown differently oriented retinal co-factors. Solid-state NMR is here the method of choice to obtain specific data to resolve this issue.
Isotope labelled retinal was incorporated into residue-selectively labelled opsin. The observation of dipolar contacts between the 13C spins enabled the authors to conclude details on the retinal orientation within the binding pocket after light activation and trapping the Meta-II state.
The authors suggest that in the first stage of the activation, the retinal beta-ionone ring triggers rotation of helix 6 and stabilizes Meta II. In the second step, C19 and C20 in the retinal polyene chain are suggested to sterically interact with Tyr191 and Tyr268, which also includes deprotonation of the retinal Schiff base.
Overall, data interpretation is sound and solid-state NMR is certainly the right approach. However, I could not follow how the authors derived a sequential step model, for which time-resolved data or at least trapping of batho, meta-I and meta-II would have been required.
Although I am familiar with the field, I would also recommend some alterations to the paper as it is difficult follow without consulting the literature and some data presentation appears selective: The materials and method section is very short. Some more details would be helpful, especially, regarding the state of the samples. Was solid-state NMR applied to frozen detergent solution or to proteoliposome samples? What is the source organism of the rhodopsin used here?
It would be helpful to include some of the 2D spectra leading to the data in Fig. 1. The authors only show 1D slices of the spectra. Fig. 1G: The authors mention a comparison to build-up curves from model compounds but it is not clear whether these data are shown here. Furthermore, spin-diffusion build-up curves can be difficult to quantify. The authors seem to present here calculated curves for certain distances. Further details should be provided.
Some explanations should be provided how the MI/MII trapping was performed and how good the trapping efficiency was.
It should be somewhere summarized how the assignment was achieved.
The first reviewer found the "manuscript to be experimentally solid" and to "powerfully combine forces of the experts in GPCR expression, retinal analogs, solid-state NMR, and FTIR spectroscopy of photoreceptor proteins".
The reviewer raised 5 major points and 42 additional points that we have addressed below. Point 1. While the experimental results are very solid and interesting, and must be published, this reviewer believes that the integrated hypothesis of a two-stage rhodopsin activation is not presented convincingly. While the proposed mechanism may be 100% correct, the presentation of this mechanism, and the way it integrates the experimental data with the vast body of the previous knowledge, need to be improved. In the present form, it is inaccessible to anyone not in the immediate narrow field, the abstract and the discussion look disconnected from the experimental results, and the whole paper looks somewhat disjointed.
In my opinion, the paper would be much stronger if the focus were shifted from the "high level integration" two-stage hypothesis to the discussion of the true orientation of retinal in Meta II in the context of associated protein changes. On the other hand, if the focus remains the same, the hypothesis should be much better integrated with the presented results and placed in context of the existing knowledge.
Response: The major concern of the reviewer is the presentation of the results. The manuscript was originally focused on defining the orientation of the retinal chromophore. However, in the course of these studies it became clear that the differences in orientation between NMR and crystallography have very different implications about the mechanism of activation, and the manuscript shifted toward a new focus.
To address the concern of the reviewer in terms of the presentation and access to the nonexpert reader, we have revised the title, abstract, introduction and discussion.

Title:
We have changed the title in order to highlight both the "retinal orientation" and the "activation mechanism" aspects of the manuscript. We believe the retinal orientation component will be of considerable interest to the visual receptor community, while the activation mechanism will have impact across the GPCR field where the details about how agonists trigger activation are lacking.

Abstract.
We have revised the abstract as suggested by the reviewer to address the concern listed below that the proposed mechanism integrates old and new data.
"We integrate these observations with previous structural and functional studies to propose a two-component mechanism for rhodopsin activation." We have also explicitly indicated that the orientation of the retinal chromophore found in Meta II differs from that in the crystal structures.  13 . This observation is surprising as a substantial amount of absorbed light energy (~35 kcal/mol 14 16,17 , consistent with electrostatic interactions having the dominant contribution to receptor activation 19 ."

Discussion.
We have revised the discussion extensively as suggested in points 20-25 below, removed the subheadings as per journal style, and edited the text to highlight the importance of these results in understanding the general mechanism of GPCR activation.
Point 2. More specifically, the following points will be unclear for most readers: i) why the retinal proximity data are collected on Meta II only (not on Meta I), but used to argue for the stage I of the triggering and ii) why and how the NMR data collected just for a few residues are used to build much more general mechanism. In other words, the experimental results should be put into the context of what is already known much more explicitly, and the new and old results should be contrasted better.
Response: We have revised the concluding paragraph in the introduction to more clearly indicate how we use NMR and FTIR to probe Meta I and Meta II. We have also revised the end of the first paragraph in the Results section to describe how select NMR distance measurements are combined with previous studies in the literature to propose a general mechanism of activation.
Pages 5 (Introduction). "In refining the orientation of the all-trans retinal in Meta-II based on solid-state NMR measurements, we propose an activation mechanism that builds on previous studies by emphasizing the changes in extracellular residues in close proximity to the retinal. We are able to follow changes in these residues by comparing differences between rhodopsin and Meta-II, or in some cases, between rhodopsin and Meta-I, which precedes Meta-II in the photoreaction pathway. To address how specific residues in close association with the retinal control the equilibrium between Meta-I and Meta-II (i.e. the final step in the reaction pathway), we take advantage of FTIR spectroscopy 20 . FTIR difference spectroscopy provides a complementary approach to NMR for characterizing the contribution of hydrogen-bonding interactions of specific amino acids to receptor activation [20][21][22] Figure 7) and Methods sections. We now show in Supplementary Figure 7 13 Cz-tyrosine, 13 Ca-glycine and containing 13 C12, 13 C20-retinal." Supplementary Figure 6. "Panel (a) presents a row through the diagonal resonance of 13  2) p. 2. abstract -see above. The main point that the proposed hypothesis integrates a lot of the old data with some of the new data presented in the paper is missing. A better placement into the context would be helpful.
Response: We have revised the abstract accordingly.
Abstract. "We integrate these observations with previous structural and functional studies to propose a two-component mechanism for rhodopsin activation." 3) p. 3, "As a result, it has been a challenge" -the logic is not clear.
Response: We have removed this sentence in the process of improving the flow and logic in the introduction. 4) p. 5, "The NMR and FTIR approaches make use of low temperature to trap the active Meta-II state" -to be fair, one can argue that NMR experiments are done on the frozen detergent micelles, which is not completely native. On the other hand, it seems that FTIR results were obtained on the HEK cell membranes (even though the methodology description is not very clear here). The difference in sample conditions between NMR and FTIR is not discussed.
Response: We now include a section in the expanded Methods to describe the differences in sample preparation and experimental conditions.  20,43 , which requires the use of time-resolved methods to follow the transition. For NMR, we convert fully to Meta-II at room temperature, but require several minutes to low-temperature trap the Meta-II intermediate, which we assume is predominately Meta-IIbH+, before it decays." " 5) p. 7, "the relative intensity of the cross peak to the Phe261" -relative to what?

The conditions for NMR and FTIR are different. For NMR, the analysis relies on complete conversion (>90%) to Meta-I or Meta-II, which is facilitated by solubilization in digitonin or DDM, respectively. Rhodopsin is monomeric in DDM and is able to activate the G protein transducin 58 . Digitonin is unusual in that its hydrophobic end is composed of a rigid spirostan steroid moiety rather than flexible fatty acyl chains. The rigid framework effectively blocks the transition from Meta-I to Meta-II 59 . For FTIR, the analysis uses difference methods, which allows one to easily shift the equilibrium between Meta-I and Meta-II by pH or temperature. At low temperatures (below ~10 °C), the Meta-I ⇔ Meta-II equilibrium reflects a two-state transition in both DDM and PC bilayers, which breaks down into a series of Meta-II substates at higher temperature 20,43 . In both DDM and PC bilayers the conversion to Meta-IIbH+ happens rapidly (millisecond-second time scale) at 20 °C
Response: We revised this sentence.
"However, the intensity of the cross peak to the Phe261 6.44 ring 13 C resonances remains approximately the same as in rhodopsin indicating that the ring does not change orientation (i.e. flip) in the conversion to Meta-II." 6) p. 7, reference to Fig. S2, actually refers to Fig. S2B. The data presented in Fig. S2A are never discussed or referred to in the main text.
Response: We have revised this sentence accordingly.
Page 7. "The position of the β-ionone ring is additionally constrained by contacts between the 13 C5, 13 C18 and the 13 C16, 13 C17 retinal resonances and residues (Met207 5. 42 and His211 5.46 ) on H5 (Supplementary Fig. 3). Furthermore, in the Meta-II-opsin and Meta-II-M257Y crystal structures, the rotation of the β-ionone ring is predicted to bring the C18 methyl group to within the DARR distance limit (~6 Å) of 13 Cζ-Tyr191 EL2 , which is not observed (Fig 1d)." 7) p. 7, "lost, consistent with an increased separation between these tyrosines and the retinal C19 carbon" -an alternative explanation would be retinal rotation, which should be mentioned and argued against once more.
Response: We have added a sentence at this point in the text. Fig. 4)." 8) p. 7, "loss of intensity of these tyrosines with the retinal C19 methyl resonance" -style Response: The sentence has been revised.  Fig. 4)." 9) p. 8, "The row through the C20 diagonal resonance yields" -here and elsewhere, it should be clearly stated that this resonance may change between Rd and Meta II and the actual positions indicated in the figures.  10) p. 9, "Tyr268 6.51 has the highest subfamily conservation" -elsewhere in the paper it says "one of the highest", also what about W265?

Page 8. "Moreover, we show that the alternative explanation, the flip of the β-ionone ring, does not occur until Meta-II decays (Supplementary
Response: Tyr268 6.51 is the second highest. Lys296 is the highest. Trp265 is highly conserved across the family A GPCRs and not just in the opsin subfamily. We added a sentence to make this point about subfamily vs family conservation clearer. Page 10. "Tyr268 6.51 has the highest subfamily conservation (97%) in the visual GPCRs after Lys296 7.43 , the site of retinal attachment, indicating that its position and interactions are critically important within the visual receptors." 11) p. 11, "This cross peak does not lose intensity as previously assigned" -style Response: The description of the cross peak intensity changes on page 11 has been rewritten and this sentence has been deleted. 12) p. 11, the discussion of Fig. 2 peak intensity changes needs to be more quantitative. E.g., what is the intensity increase for the 268/122 crosspeak?
Response: We have added selected rows from the contour plots in Figure 2 to the figure in order to clearly show the intensity changes. The most relevant rows correspond to the Tyr-Gly cross peaks. (We do not have a 268/122 cross peak; rather the cross peak intensity that the reviewer is referring to is likely 268/188).
We have modified the figure legend to discuss the intensity differences in these rows.
Page 30: "Above panel (a) are shown rows through the Tyr-Gly crosspeaks. The rows better illustrate the intensity change occurring in the Tyr268 6.51 -Gly188 EL2 peak upon activation. The observation that the Tyr178 EL2 -Gly114 3.29 does not change intensity is consistent with the lack of influence of the Y178F mutation upon the Meta-I -Meta-II transition (see Fig. 4g)." 13) p. 11, "Both increase to > 6 Å" -style Response: The sentence has been revised.
Page 11. "Both Tyr-Cys distances increase to >6 Å in the opsin crystal structure 4 ." 14) p. 12, "Tyr268 6.51 on TM helix H6 is strongly hydrogen bonded to Glu181" -indicate according to which structure Response: In the supporting Tables, we have listed the crystal structure distances that are likely to be of interest to the reader. The Tyr268-Glu181 distances were not previously included. We have now listed the O…O distances between the Tyr268 side chain that the Glu181 carboxyl side chain in Table S5 and made a note of this in the main text.
Page 12. "Tyr268 6.51 on TM helix H6 is strongly hydrogen-bonded to Glu181 EL2 on the β3 strand of EL2 and to Tyr191 EL2 on the β4 strand of EL2 (Fig. 3d, Supplementary Table 5)." 15) p. 12, "and that Tyr191EL2 has shifted relative to Glu181EL2" -not clear where this came from, as E181 has not been observed in this paper. Needs better explanation.
Response: This sentence has been revised. Response: The schematic in Figure 5 was intended for this purpose. We have included additional views of the region containing Met288, Tyr191, Tyr268 and Glu181 in Supplementary Figure 6 that should help visualize the changes.
19) p. 14, "The observation that Tyr191 EL2 and Tyr268 6.51 strongly stabilize the Meta-I state suggests that in crystal structures of opsin and/or Meta-II, the inactive hydrogen-bonding network reforms in this region of the receptor upon the decay of Meta-II to opsin" -needs better explanation Response: This sentence has been moved to the Discussion section and revised.
Pages 18. "The net effect of PSB deprotonation and rearrangement of the hydrogen-bonding network involving EL2 is a shift of Tyr191 EL2 away from H6. We propose that this motion allows the extracellular end of H6 to pivot inward. In the visual pigments, Tyr191 EL2 has a high level of sequence identity (61%), and an overall level of conservation of 83% as either tyrosine or tryptophan. Interestingly, even though Tyr191 EL250 or Tyr268 6.51 contribute to Meta-I stability (Fig. 4), mutation of either these residues results in a substantial drop in G protein activation 38,50 . This dual influence of Tyr191 EL2 and Tyr268 6.51 on Meta-I stability and Meta-II activity is consistent with one set of hydrogen-bonding interactions stabilizing Meta-I and a second set stabilizing Meta-II (Fig. 5) 24) p. 16, "results in motion of the retinal PSB proton away from the stabilizing interaction with its counter-ion Glu113" -a bit unorthodox, as normally people speak about the PSB counterion rather than proton counterion.
Response: The issue here is that the nitrogen of the PSB actually bears partial negative charge, while the positive charge on the retinal PSB contributed by protonation of this nitrogen is distributed on various atoms on the retinal and lysine side chain. The largest partial positive charge is on the Schiff base proton. Hence, the relative orientation of the  Fig. 1 legend -PDB accession should be given; positions of the diagonal resonances used should be indicated; "scaled to the C12-C20 cross peaks" -specify how; Response: The figure legend has been revised. We now describe the scaling of C12 and C20 cross peaks in Supplementary Fig. 2.
"Strong cross peaks are also observed between the 13 C20 retinal resonance and the 13 Cζresonance of Tyr268. The weak cross peak between 13 C12 and 13 Cζ-Tyr268 (relative the C12-C20 "internal control") indicates a longer internuclear distance. The intensity of the C12-C20 "internal control" allows us to scale the 13 C12- 13 Cζ-Tyr268 and the 13 C20 -13 Cζ-Tyr268 cross peaks relative to each other." 28) Fig. 1H -Signal-to-noise is poor resulting in some negative bands (phasing?). May be comment on that?
Response: We re-examined the spectra and the "negative peak" is just an artifact of poor signal-to-noise. We have other data in which there is not a negative peak, but these data were obtained by scanning rows to get the maximum C14 and C15 cross peak intensity are the best. The data sets on rhodopsin and Meta II were obtained and scaled to be comparable. Fig. 3 legend -"are shown of rhodopsin minus Meta-I (orange) and rhodopsin minus Meta-II (black) of rhodopsin" -style

29)
Response: The sentence has been revised. 37) Suppl. p. 12 -"absence of a cross peak between the retinal chromophore and 13 Cζ-labeled tyrosine in Meta-II" -not true in general, should specifically name the retinal part discussed Response: Corrected.
Page 8 (now Supplementary Fig. 4). "This conclusion was based in part on the absence of a cross peak between the 13 C18 methyl group on the β-ionone ring of the retinal chromophore and 13 Cζ-labeled tyrosine in Meta-II (Fig. 1c)." 38) Suppl. Fig. S6 -normalization of the C5-C18 cross peak is confusing, as its amplitude decreases Response: We have revised the text to better convey the fact that the C18-Tyr intensity is increasing relative to the decaying Meta II signal from the C5-C18 cross peak.
Page 9 (now Supplementary Fig. 4 Supplementary Fig. 4). "The high frequency of the 13  The reviewer lists several minor issues that need correction prior to publication.
1. Figure 1 caption is confusing: Does Figure 1B show crystal structure of dark state or Meta-II intermediate. Also cytoplasmic and exterior membrane surface should be labeled in Figure 1B. Note that Tyr191 does not appear to be the closest residue to retinal C18. If Figure 1B is the dark state then this is understandable but in that case one of the Meta II forms of the crystal structure should also be shown.
Response: Figure 1b is the dark-state crystal structure. We have now added the PDB ID in the figure legend. We have also labeled the two surfaces and revised the figure caption. We show the comparison with the Meta-II crystal structure in Supplementary Fig. 1. We believe the new figure in Supplementary Fig. 1 will help the reader in visualizing the differences between the retinal orientation derived from the NMR measurements and observed in the crystal structures.
2. Page 6: It should be noted that the referred to peak at 21.6 ppm is not shown.
Response: The 21.6 ppm resonance corresponds to the C18 diagonal resonance. This is not shown in the portion of the spectrum displayed. We have added a supporting figure (new Figure S2) that explains how where the rows and cross peaks come from in a full 2D solidstate NMR spectrum. This should help the reader that is not familiar with this type of data.
3. Figure 1C: In bottom row of DARR data add lines to show the Phe261-C18 cross peaks as done for DARR row above.
Response: Lines have been added as suggested.
4. Page 7, paragraph 2: Thr118 is not shown in Figure 1B but mentioned in text. The authors should consider adding this to Figure. Response: Thr118 has been added to Figure 1b.
5. Figure 2B: The label Ile189 is partially clipped at top of font.
Response: We were not able to find the label that the reviewer is referring to. We have double-checked all of the figures to ensure that they are not being clipped.
6. Figure 3: Crystal structure is mislabeled C and not D.
Response: Corrected 8. Figure 4: It would be easier to compare these plots if the pKas (inflection points) were indicated on the curves. This is especially true since x-axis pH scale is not reproduced on top set of panels.
Response: We added the pH scale on the top set of panels and added one vertical dashed line in each panel at the pKa in order to guide the eye.
9. Discussion: An earlier study which involved FTIR difference spectroscopy of isotope labeled rhodopsin suggested a role for Tyr268 in rhodopsin activation as well as helix-6 reorientation. This work should be mentioned and referenced to in Discussion (DeLange, et al. (1998) Tyrosine structural changes detected during the photoactivation of rhodopsin. J Biol Chem 273, 23735-23739).
Response: We have added this reference and a short discussion of the work.
Page 17. "These results are consistent with an earlier FTIR study suggesting a role for Tyr268 and H6 reorientation in rhodopsin activation 44 . " 10. Supplementary Figure S5: Panel D is not described. For example, which state does purple colored structure refer to?
Response: Additional explanation is added to the figure legend.
Page 19 (now Supplementary Fig. 7). "The purple cylinders are a cartoon representation of the mechanism proposed here in which deprotonation and the associated changes in the extracellular hydrogen-bonding network allow the intracellular end of H6 to pivot inward and the extracellular end of H6 to pivot outward." solution or to proteoliposome samples? What is the source organism of the rhodopsin used here?
Response: Methods section has been expanded. The samples were frozen in DDM detergent and the source organism was cows (bovine rhodopsin).
Page 19. "Expression and purification of 13  3. It would be helpful to include some of the 2D spectra leading to the data in Fig. 1. The authors only show 1D slices of the spectra.
Response: We added a supporting figure (Supplementary Fig. 2) showing the full 2D plot for the experiment using 13 Cζ-Tyr and 13 C12,20-retinal. This figure is intended to illustrate where the rows originate from in Figure 1 and where the contour plots showing originate from in Figure 2.
4. Fig. 1G: The authors mention a comparison to build-up curves from model compounds but it is not clear whether these data are shown here. Furthermore, spin-diffusion build-up curves can be difficult to quantify. The authors seem to present here calculated curves for certain distances. Further details should be provided.  16,17,29,32,56,57 ." 5. Some explanations should be provided how the MI/MII trapping was performed and how good the trapping efficiency was.