Aedes aegypti Piwi4 Structural Features Are Necessary for RNA Binding and Nuclear Localization

Round 1

Reviewer 1 Report

This manuscript by Williams et al., explores the Piwi4 protein in the mosquito Aedes aegypti. Specifically, it examined the binding affinity of Piwi4 to piRNA, the demonstrated that mutations in key amino acids in the binding domain altered the structural backbone of the protein, impacting its capacity to bind to piRNAs, and finally, they demonstrated that the protein was localized in both the nucleus and cytoplasm and identified a putative nuclear localization signal. Overall, this is an interesting and technically sound manuscript. While this work doesn’t provide a huge advance in the biology of Ae. aegypti, it does provide a detailed examination of a protein group that has expanded in Aedes, relative to other Dipterans. I do not have any major concerns about this work, and I think it would be of interest to the readers of the International Journal of Molecular Sciences.

A few minor issues:

Scientific names are not italicized throughout the text.

In figure 1D the arrows directing readers to the key binding residues need to be more carefully placed. It is difficult to tell which residues the arrows are pointing to in some cases.

In Figure 2 and 3 remove the 1X10-5 concentration label as it appears this concentration was never tested. Also in these figures it would be helpful to keep the Y-axis scales the same.

Line 381-382: is Drosophila not an arthropod?

In the antibody test the size of the recombinant protein is markedly different from the endogenous protein based on the MW ladder (supplemental fig 5 and supplemental, supplemental figure 5). Is the endogenous protein expected to be larger than the recombinant? If yes this should be clarified? The authors state that “Mass spectrometry analyses further confirmed that Piwi4-specific peptides were present at the same location on a corresponding SDS/PAGE gel slice”, but wouldn’t that be true whether or not the antibody recognized the target protein as the size would correspond to the gel fragment? There is also a considerable amount of antibody binding to other proteins, raising the question of whether the authors are really seeing the piwi4 protein in the IFA assays. Was preimmune serum collected from the mice to possibly eliminate some of the bands? Were the IFA assays repeated with serum from a second mouse for confirmation (eg. the mouse serum from lane 5 (supp, supp fig 5) could act as confirmation of the lane 4 serum even though more background binding is observed).

Author Response

We thank the reviewers for their careful attention to detail and productive comments and suggestions. We have responded below to each comment in a point-by-point manner in red. We have also made the corresponding changes in the manuscript using the track features function on Word.

Comments and Suggestions for Authors

This manuscript by Williams et al., explores the Piwi4 protein in the mosquito Aedes aegypti. Specifically, it examined the binding affinity of Piwi4 to piRNA, the demonstrated that mutations in key amino acids in the binding domain altered the structural backbone of the protein, impacting its capacity to bind to piRNAs, and finally, they demonstrated that the protein was localized in both the nucleus and cytoplasm and identified a putative nuclear localization signal. Overall, this is an interesting and technically sound manuscript. While this work doesn’t provide a huge advance in the biology of Ae. aegypti, it does provide a detailed examination of a protein group that has expanded in Aedes, relative to other Dipterans. I do not have any major concerns about this work, and I think it would be of interest to the readers of the International Journal of Molecular Sciences.

A few minor issues:

Scientific names are not italicized throughout the text.

All species names have been italicized throughout the manuscript.

In figure 1D the arrows directing readers to the key binding residues need to be more carefully placed. It is difficult to tell which residues the arrows are pointing to in some cases.

We thank the reviewer for pointing out this confusion. We have explicitly added the amino acid residue number from the alignment that are known to bind RNA in the figure legend. We also redid the figure so that the arrows are closer to the amino acid letters.

In Figure 2 and 3 remove the 1X10-5 concentration label as it appears this concentration was never tested. Also in these figures it would be helpful to keep the Y-axis scales the same.

We agree with the reviewer that the 1x10^-5 M concentration was never tested, and that the X-axis should be appropriately scaled to reflect the concentrations tested. We have made this correction to our steady-state graphs. We have also scaled the Y-axes to be the same across sample types in each experiment.

Line 381-382: is Drosophila not an arthropod?

Yes, the reviewer is correct that Drosophila is in the Arthropoda phylum, so we have added “arthropod vector” to clarify.

In the antibody test the size of the recombinant protein is markedly different from the endogenous protein based on the MW ladder (supplemental fig 5 and supplemental, supplemental figure 5). Is the endogenous protein expected to be larger than the recombinant? If yes this should be clarified?

We thank the reviewer for this observation. Indeed, the recombinant AePiwi4 full length protein (Supplemental Figure 5, lane 2) is ~100 kDa while the endogenous AePiwi4 full length protein runs on an SDS-PAGE gel ~130 kDa (Supplemental Figure 5, lane 3). To make this clearer, we have added in additional annotations for the molecular markers to Supplementary Figure 5. This discrepancy may be due to post-translational modifications to the endogenous protein, such as glycosylation (there are two predicted glycosylation sites: N306 and N820), that can impact how the protein migrates on the gel. As stated in the manuscript, Piwi4-specific peptides were detected in a corresponding gel slice where mosquito tissue had been loaded, so we are confident that, despite this difference in how the recombinant and endogenous AePiwi4 proteins runs, that AePiwi4 is present at this molecular weight in the gel.

The authors state that “Mass spectrometry analyses further confirmed that Piwi4-specific peptides were present at the same location on a corresponding SDS/PAGE gel slice”, but wouldn’t that be true whether or not the antibody recognized the target protein as the size would correspond to the gel fragment?

We agree with the reviewer that AePiwi4 will be present at this location on a corresponding SDS/PAGE gel slice regardless of whether the antibody bound the protein. The LC/MS experiment was simply performed to show that the recognized band is at the same location on the gel of where endogenous AePiwi4 is, as described in the answer previously.

There is also a considerable amount of antibody binding to other proteins, raising the question of whether the authors are really seeing the piwi4 protein in the IFA assays. Was preimmune serum collected from the mice to possibly eliminate some of the bands? Were the IFA assays repeated with serum from a second mouse for confirmation (eg. the mouse serum from lane 5 (supp, supp fig 5) could act as confirmation of the lane 4 serum even though more background binding is observed).

The reviewer brings up a critical observation. We generated the AePiwi4 polycloncal antibodies by immunizing mice with recombinant AePiwi4 PAZ. PAZ is a conserved domain across Argonaute proteins, so it is possible the AePiwi4 PAZ pAbs bind to other Arognaute proteins at their PAZ domains, including to the other A. aegypti Piwis. Indeed, in the cytoplasmic fraction of A. aegypti ovaries, AePiwi4 PAZ pAbs also recognize proteins at a much higher molecular weight than AePiwi4 (>250 kDa, Figure 5C, lane 1). To the authors’ knowledge, there are no A. aegypti Piwis or Agos this large. However, AePiwi4 does form protein complexes with other Piwis, so it is possible that this band is protein aggregates that were not completely reduced in the tissue samples.

Nevertheless, western blots (Figure 5 & Supplemental Figure 5) were performed to confirm that the AePiwi4 PAZ antibodies recognized recombinant and endogenous AePiwi4. Our subsequent studies identified a functional nuclear localization signal in AePiwi4 (Figure 6C-E) that corroborated the western blot data. Although western blot and IFAs are different techniques, control IFAs using mice serum that had been immunized with Magic Mouse adjuvant alone were used to ensure minimal non-specific binding.

Reviewer 2 Report

The study provides insight into the role of the Piwi 4 protein of Aedes aegypti, which is involved in antiviral immunity and embryonic development of the mosquitoes. The study is carefully done and the manuscript is generally well written, but needs some editorial inprovement.

• Keywords: delete "Piwi" because Piwi 4 is alreadey mentioned before
• Line 96 and others: all species names have to be written in italics, in the text, in figure legend and in the references
• Line 168 and others: insert a space between numbers and units
• Lines 196 and 230: what means "error"? SD or SEM as in other figures?
• Line 233: I could not find "Ri" in the table
• Line 247: a spectrum
• Line 304 and others: use the acronym D. melanogaster after first mentioned; dto. for Aedes aegypti
• Line 519 and others: give names of chemicals in lowercase letters
• Line 546 and others: what means "maximal speed"?; give the correct g value; m is not a correct acronym for minutes (min)
• Lines 547/548: insert commas between the various chemicals
• Line 566 and others: use either l or L for liter in the entire manuscript
• Line 568: to present centrifugation steps in rpm instead of g values is an absolute "no go" for biochemists
• Line 626: delete one bracket
• References: check the references for a uniform style of presentation according to IJMS author*s instructions.

Author Response

We thank the reviewers for their careful attention to detail and productive comments and suggestions. We have responded below to each comment in a point-by-point manner in red. We have also made the corresponding changes in the manuscript using the track features function on Word.

Comments and Suggestions for Authors

The study provides insight into the role of the Piwi 4 protein of Aedes aegypti, which is involved in antiviral immunity and embryonic development of the mosquitoes. The study is carefully done and the manuscript is generally well written, but needs some editorial inprovement.

• Keywords: delete "Piwi" because Piwi 4 is alreadey mentioned before
• Although we agree with the reviewer that having “Piwi” and “Piwi4” as keywords may sound redundant, we believe including both will increase viewership. “Piwi” may refer to a gene, a protein, a subfamily of proteins, or a protein domain. Piwi4, on the other hand, is a Piwi protein specific to Ae. aegypti. We believe including Piwi as a keyword may attract relevant attention to those that study Piwi proteins outside of the Ae. aegypti community.
• Line 96 and others: all species names have to be written in italics, in the text, in figure legend and in the references
• All species names have been italicized throughout the manuscript.
• Line 168 and others: insert a space between numbers and units
• Spaces have been added between numbers and units.
• Lines 196 and 230: what means "error"? SD or SEM as in other figures?
• Thank you to the reviewer for pointing out this omission. In what are now lines 196 and 233, we have clarified that “red bars indicate mean and standard deviation for R_max values.”
• Line 233: I could not find "Ri" in the table
• We thank the reviewer for pointing out this omission and for allowing us to clarify how the RI value was used to calculate the disassociation constants in steady-state affinity. Here, the RI is the bulk refractive index contribution that measures changes in RIs outside of the surface of the sensor due most likely to buffer changes. To calculate equilibrium disassociation constants from steady-state affinities, the RI is the offset, or the response at zero analyte concentration. The offsets are therefore relative to the individual samples and cannot be compared across sample types in the table. To clarify this confusion, we have used “offset” instead of “RI” and have defined this variable in the Methods section.
• Line 247: a spectrum
• This grammatical change has been made in the manuscript.
• Line 304 and others: use the acronym D. melanogaster after first mentioned; dto. for Aedes aegypti
• We agree with the reviewer that the acronyms should be consistent. We therefore changed Ae. aegypti to A. aegypti throughout the manuscript.
• Line 519 and others: give names of chemicals in lowercase letters
• 0.1 mM isopropyl β- d-1-thiogalactopyranoside (IPTG) in what is now line 591 has been changed to lowercase letters.
• Line 546 and others: what means "maximal speed"?; give the correct g value; m is not a correct acronym for minutes (min)
• The maximum speed has been explicitly added (“4,000 x g”), and all acronyms for minutes have been changed from “m” to “min.”
• Lines 547/548: insert commas between the various chemicals
• Commas have been added between “20 mM Tris-HCl, 150 mM NaCl, pH 7.4”
• Line 566 and others: use either l or L for liter in the entire manuscript
• We agree with the reviewer that this should be consistent, and we changed all “ml” to “mL.”
• Line 568: to present centrifugation steps in rpm instead of g values is an absolute "no go" for biochemists
• We have converted 47,000 RPM to 100,000 x g here.
• Line 626: delete one bracket
• We appreciate the reviewer pointing out issues in our use of the parentheses for the mean residue ellipticity equation. We have corrected this problem.
• References: check the references for a uniform style of presentation according to IJMS author*s instructions.
• We have double checked the references for appropriate style and have made all necessary changes.