Avalilação PREreview de Cryo-tides: synthetic peptides for cryo-EM grid preparation of multi-subunit protein complexes sensitive to the air-water interface

Peer review of preprint: Cryo-tides: synthetic peptides for cryo-EM grid preparation of multi-subunit protein complexes sensitive to the air-water interface by Young R., McGeoch A.J.S., Jessop M., Greber B.J., and Alfieri C.

Overall Assessment:

The authors propose the use of short peptides, which they name cryo-tides, as an additive for cryogenic electron microscopy (cryo-EM) samples to decrease sample interactions with the air-water interface (AWI). Interactions of proteins, particularly complexes, with the AWI can be a major bottleneck in cryo-EM structure determination with significant research being done to try and overcome this problem. The authors designed 4 different 21 amino-acid peptides, 2 with a mix of basic and hydrophobic residues (BH & SBH), 1 with acidic and hydrophobic residues (AH), and 1 poly-lysine peptide (PL). They used single particle cryo-EM and cryogenic electron tomography (cryo-ET) to determine the effects cryo-tides had on cryo-EM analysis of the anaphase-promoting complex (APC/C) and CDK11:cyclin L:SAP30BP complex (CDK11 complex). The authors clearly show how the cryo-tides were successful in improving particle distribution and the percentage of intact complexes for both protein complexes, and that datasets with different cryo-tides can be combined to give improved coverage. How these peptides function and their wider applicability of this technology was not clearly defined, but is outside of the scope of this preprint and may come as this method is applied to other systems.

Strengths: The cryo-EM and cryo-ET analyses in this preprint are well described and were rationally performed. The results clearly show the improvement that the three main cryo-tides had on improving particle completeness and viewing angle distribution, likely as a result of decreasing AWI interactions. The methodology described, particularly the combination of multiple cryo-tide datasets, is an important result targeting a major pain-point in cryo-EM sample preparation.

Limitations: The overall scope of this study is limited, making it hard to access the broad applicability of this methodology. The small number of cryo-tide sequences and conditions tested make it difficult to determine what factors (e.g. % polar/charged vs hydrophobic residues, peptide length, peptide concentration, use of multiple peptides in a single sample) may affect how these cryo-tides function and how other users may optimize the method for their own experiments. Additionally, the small number of proteins tested means the broader usability of this method is not clear.

Major Points:

1. On page 3, it is mentioned that chemical additives like detergents may be directly binding with the protein on the grid and stabilizing the sample that way. Binding is also a possibility for the cryo-tides, which is not mentioned or investigated. Biochemical binding experiments and/or noting whether or not there is density for the cryo-tides within the maps would indicate whether the observed effects may be a result of direct binding to the proteins. This additional information would help readers who may be interested in adapting this method know whether cryo-tide binding is an optimizable characteristic.

2. Table 1 and figure S8 both show that the cryo-tides generally lead to a substantial decrease in the percentage of accepted micrographs (91% ctrl APC/C vs 80% SBH and 62% AH, and 85% ctrl CDK11 complex vs 82% cryo-tide). This effect is not commented on in the preprint. Describing why these micrographs are rejected and whether or not this is likely a result of the cryo-tides, could help future users optimize this method and use it for their own systems.

3. On page 6, in the conclusion paragraph for the results sections, the authors state that the hydrophobic component of the peptides is essential for protective action. Although this feature is indicated by the current data, the small number of peptides and proteins tested likely doesn’t establish this as a universal principle. Without thorough analysis of peptides with different lengths, polarities, charges, and other biochemical attributes, definitive determination of what is and is not needed may not be appropriate. I would suggest rephrasing this conclusion as potential rather than definitive, which may be more in-line with the supporting evidence provided.

4. How the cryo-tides compare to the other preparation strategies described in the introduction that can also decrease AWI interactions is not described. It would strengthen the paper to include a comparison of some of the statistics of the APC/C cryo-tide structures (e.g. % intact particles, 3DFSC sphericity, etc.) with the previously collected APC/C structure from a grid with a thin carbon film. This addition would give context to how this methodology compares and help readers determine which may be most appropriate for their experiments.

5. As a final conclusion on page 23, the authors mention that these studies may act as a starting point for future cryo-tide investigation and design, but do not indicate what these future experiments may be. Although systematic analyses would be required to definitely determine what these peptides could be optimized and what features may be dependent on the protein sample, a short section suggesting potential starting points may be helpful. Providing some potential starting experimental variables such as % polar/charged residues, cryo-tide size, and cryo-tide concentration may be helpful especially if other researchers do not immediately see improvement in their data using the cryo-tides as described here. This addition would give researchers trying to utilize this method for their own experiments a starting point, and make it clear what some future directions may be.

Minor Points:

1. Figure 2 shows the individual components of the APC/C complex but does not clearly indicate the named modules that are used to determine whether or not the complex is complete. Making proteins that are in the same module similar colors or highlighting the modules in some other way would make it easier for readers who are unfamiliar with the APC/C complex to identify the modules.

2. Figure 3 states that the ab initio maps are displayed at the same threshold values using ChimeraX, but this same indication is not given for other figures showing the maps such as figures 4, 5, and 9. Mentioning how these structures were visualized, either in the figure captions or in a “structure visualization” methods section, and explicitly stating the threshold value used would help readers who may want to look at these structures themselves in 3D.

3. In the “Grid preparation” method section, the protein concentrations mentioned are 1 μM for APC/C and 0.4 mg/mL for the CDK11 complex. To facilitate protocol adaption, using the same concentration unit (or both) for both proteins here would make it easier to compare the two concentrations and adapt the methodology for other proteins.

4. As described in the “Cryo-EM Data Processing” methods section, the SBH APC/C dataset was processed differently than the other cryo-tide containing datasets. Describing why this was the case would make it clearer for people trying to utilize this method for their own experiments.

5. The supplementary figures are inconsistently labeled. The first supplemental figure is numbered 10, continuing from the last number of the main text figures. However, the in-text citations for the supplementary figures start at supplementary figure 2.

Disclosure:

I reviewed this preprint as part of the Transparent and Accountable Peer Review training program. This report was written with the support of my research mentor. I have significant cryo-EM experience with protein complexes and trying to overcome AWI interactions and preferred orientations, but not with the specific complexes utilized in this preprint. The contents of this review were inputted into an LLM to assess sentence structure and tone, but not to review the preprint itself. I declare that I have no competing interests in relation to this manuscript.

Competing interests

The authors declare that they have no competing interests.

Use of Artificial Intelligence (AI)

The authors declare that they did not use generative AI to come up with new ideas for their review.