« premier » 3 lignes plus haut et « première » ici : y a-t-il moyen d'éviter la répétition ?

presque

quinzaine

pas tout à fait clair : le texte continue de s'effacer, mais le poème se poursuit seulement en images et sons ? Je suggère de reformuler plus clairement.

une alerte qui nous invite à ne pas oublier l'image apparaît

I found this passage disturbing in a few ways. When a person has their leg amputated, has a mastectomy, or any sort of change to their body, self concept, and their function, a part of their "will" will alter because their body does not work in the same way anymore. Yes, a person can be reflective and not let this damper their life...but I feel like directly after such a major change to one's body, they will have to re-learn how to live certain parts of their life whether this be walking, wardrobes, driving, talking or any daily activities of life... Self-concept and how a person sees themself and their opinion is a major theme in Stoicism, and disease/impediments can easily change a person's concept of their own individuality and body. There will be good and bad days, mindset is very important to healing and making progress in life.

I thought this was an interesting phrase that talked about possessions and validation. The elated horse thinks he is beautiful, meaning that he is gathering validation from internal happiness and character. But the person who says "I have a beautiful horse" is using appearances to find happiness, instead of finding it within themselves. This paragraph is important to recognize the vitality of what we truly think is necessary in our lives, and that the only thing that truly belongs to us is ourselves and our opinions. That person will never truly own the horse, and therefore the elation of owning the horse will never be completely fulfilling.

I found this passage enlightening. I feel like as humans when something negative happens to us, we want to blame the people around us in order not to take responsibility and to continue to think of ourselves in a positive light instead of looking inward. I think that one of the whole points of stoicism is self-awareness and reflection on our own actions/opinions and this quote encapsulates that idea. Rather than blaming someone else for our negative emotions that they had no control over, rather look deeper into the situation and our own attitude.

typo fixture ＆ 意味が伝わりにくいと感じました。

代案（ちょっと補足）

fixtureにより、テストごとにインスタンスを作らなくても1つのインスタンスだけで処理できます。

ちょっとわかりにくいと感じました。

原文

This usually happens in the early stages of a project when desired behavior is still being sorted out, and no one is using your code yet.

代案

このようなことは通常、プロジェクトが初期の段階で望ましい動作を整理している最中であり、かつ誰もまだ使い始めていないときに発生します。

typo を実行

文の前半とのつながりがよくないと感じました。

代案１（原文に忠実） テストが再度成功しました。これは元の機能が損なわれていないということを意味します。

代案２（ちょっと意訳） これでテストが再度成功し、元の機能が損なわれていないということが確認できました。

前の節（テストに成功する）、続く説明と揃えて「テストに失敗する」にしてはどうでしょう

more の語感が抜けていると思います。

代案

より多くの人がプログラムを使い始めるときに

I will be adding these questions to my practice of self-reflection. I think these questions are poignant. I will also look for a way to incorporate them into self-reflections I ask my students to provide.

I have been using Canvas for collaboration, but I believe that incorporating Hypothesis will allow my students to collaborate in a way that is helpful and meaningful to their learning.

As a result of this course I am "tak[ing] a hard look" at the tools availabile to me and questioning if I am using the right ones and am I using the tools I do use to the best advantage of my students.

I do not believe that I have been doing my best to energize learning in my students through my use of digital tools. I am learning new ways of elevating learning and energizing my students by using the digital tools I have been using better, and by incorporating new digital tools.

Case: Patient #42, female, Korean

DiseaseAssertion: UCD/OTCD

FamilyInfo: N/A

CasePresentingHPOs: HP:0011463, HP:0001987, HP:0003218

CaseHPOFreeText: Childhood onset, hyperammonemia, oroticaciduria

CaseNOTHPOs:

CaseNOTHPOFreeText:

CasePreviousTesting: Genomic DNA was extracted from peripheral blood leukocytes. A total of 10 coding exons and exon–intron boundaries of the OTC gene were amplified by PCR with customized primers. PCR products were directly sequenced with the same primers . Sequencing results were compared with the established human OTC sequences(NM_000531.5). Multiplex ligation-dependent probe amplification analysis was performed for patients in whom no OTC mutations were identified by direct sequencing using the OTC MLPA kit.

Supplemental Data: Table 1, no range was given for blood ammonia concentration, range given in the tables for glutamine and urine orotate is slightly different than the one in the results paragraphs.

Variant: NM_000531.6:c.958C>T(p.Arg320*)

ClinVarID: 97371

CAID:CA285809

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #44, female, Korean

DiseaseAssertion: UCD/OTCD

FamilyInfo: variant in mother (father not tested)

CasePresentingHPOs: HP:0011463, HP:0001987, HP:0003218, HP:0003217

CaseHPOFreeText: Childhood onset, hyperammonemia, oroticaciduria, hyperglutaminemia

CaseNOTHPOs:

CaseNOTHPOFreeText:

CasePreviousTesting: Genomic DNA was extracted from peripheral blood leukocytes. A total of 10 coding exons and exon–intron boundaries of the OTC gene were amplified by PCR with customized primers. PCR products were directly sequenced with the same primers . Sequencing results were compared with the established human OTC sequences(NM_000531.5). Multiplex ligation-dependent probe amplification analysis was performed for patients in whom no OTC mutations were identified by direct sequencing using the OTC MLPA kit.

Supplemental Data: Table 1, mother is a carrier, no range was given for blood ammonia concentration, range given in the tables for glutamine and urine orotate is slightly different than the one in the results paragraphs.

Variant: NM_000531.6:c.799_800insA (p.Ser267Lysfs*26)

ClinVarID: N/A

CAID:CA2695233327

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

As you read this short story consider what was happening in the world at the time this story was written - 1966. If you are unsure do some research. You know that the author is female and her main character, Connie, is female. What clues in the story connect with what is happening in the world and how that affects what is happening in the story?

Case: Patient #30, male, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo: no family history of the disease

CasePresentingHPOs: HP:0003593, HP:0001987, HP:0003218

CaseHPOFreeText: infantile onset, hyperammonemia, oroticaciduria

CaseNOTHPOs:

CaseNOTHPOFreeText: No neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, drug treatment(L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate)

Variant: NM_000531.6:c.929_931del(p.Glu310Valfs*45)

ClinVarID: 858012

CAID:CA916083888

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #61, female, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo:

CasePresentingHPOs: HP:0011463, HP:0001987, HP:0003218, HP:0003572

CaseHPOFreeText: childhood onset, hyperammonemia, oroticaciduria, low plasma citrulline

CaseNOTHPOs:

CaseNOTHPOFreeText: neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, drug treatment(L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate)

Variant: NM_000531.6:c.868-1G>C

ClinVarID: N/A

CAID:CA412725475

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #52, female, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo: family history of the disease, variant in probands mother and father

CasePresentingHPOs: HP:0011463, HP:0001987, HP:0003218, HP:0003572

CaseHPOFreeText: childhood onset,, hyperammonemia, oroticaciduria, low plasma citrulline

CaseNOTHPOs:

CaseNOTHPOFreeText: neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, inherited mutation, drug treatment(L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate), low protein diet treatment, and continuous veno venous hemodialfiltration

Variant: NM_000531.6:c.703C>T

ClinVarID: N/A

CAID: CA412721652

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #20, male, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo:

CasePresentingHPOs: HP:0011463, HP:0001987, HP:0003218

CaseHPOFreeText: childhood onset, hyperammonemia, oroticaciduria

CaseNOTHPOs:

CaseNOTHPOFreeText: No neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, drug treatment (L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate), deceased

Variant: NM_000531.6:c.794G>A(p.Trp265*)

ClinVarID: N/A

CAID:CA412722994

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #60, female, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo: family history of the disease

CasePresentingHPOs: HP:0011463, HP:0001987, HP:0003218, HP:0003572

CaseHPOFreeText: childhood onset,, hyperammonemia, oroticaciduria, low plasma citrulline

CaseNOTHPOs:

CaseNOTHPOFreeText: neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, inherited mutation, drug treatment(L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate), low protein diet treatment, and continuous veno venous hemodialfiltration

Variant: NM_000531.6:c.718-1G>A

ClinVarID: N/A

CAID: CA412722112

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #58, female, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo: variant pin proband's mother and father CasePresentingHPOs: HP:0003593, HP:0002038, HP:0001987, HP:0003218, HP:0003572

CaseHPOFreeText: infantile onset , protein avoidance, hyperammonemia, oroticaciduria, low plasma citrulline

CaseNOTHPOs: N/A

CaseNOTHPOFreeText: neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purifed and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplifcation analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, low protein diet treatment and drug treatment(L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate), neurological damage

Variant: NM_000531.6:c.540+265G>A

ClinVarID: 449382

CAID: CA658658977

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #3, male, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo: Family history of the disease, variant in probands mother and father

CasePresentingHPOs: HP:0003623, HP:0001987, HP:0003218

CaseHPOFreeText: neonatal, hyperammonemia , oroticaciduria

CaseNOTHPOs: N/A

CaseNOTHPOFreeText: No neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, no therapy received, mutation inherited, family history, deceased

Variant: NM_000531.6:c.579G>A

ClinVarID: N/A

CAID: CA412725369

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Case: Patient #59, female, Chinese

DiseaseAssertion: UCD/OTCD

FamilyInfo: variant in proband's mother and father

CasePresentingHPOs: HP:0003621, HP:0001987, HP:0003218, HP:0003217

CaseHPOFreeText:juvenile onset, hyperammonemia , oroticaciduria, hyperglutaminemia

CaseNOTHPOs: N/A

CaseNOTHPOFreeText: No neurological damage

CasePreviousTesting: gDNA extracted from peripheral blood leukocytes. PCR all coding exons and exon–intron boundaries of the OTC gene using 9 pairs of synthetic oligonucleotide primers, and the primer sequences and annealing temperature. PCR products were then purified and bidirectionally sequenced. The library was sequenced using Illumina HiSeq4000 and generated 150 bp paired-end reads. Data analysis was performed as previously described [Sun Y, Hu G, Liu H, Zhang X, Huang Z, Yan H, et al. Further delineation of the phenotype of truncating KMT2A mutations: the extended Wiedemann–Steiner syndrome. Am J Med Genet A. 2017;173:510–4.]. Multiplex ligation-dependent probe amplification analysis was performed for samples in which Sanger sequencing or WES failed to detect any disease-causing mutation.

SupplementalData: Table 3, drug treatment(L-arginine, L-Citrulline, sodium benzoate, and sodium phenylbutyrate)

Variant: NM_000531.6:c.664-1G>A

ClinVarID: 97288

CAID: CA224737

gnomAD:

GeneName: OTC (ornithine transcarbamylase)

Certainly! Here's a simplified explanation and notes about the Response() class in Django REST framework:

Response Class in Django REST Framework

• Purpose: The Response() class in Django REST framework is used to send data back to clients in various formats, such as JSON or HTML, based on what the client requests.

• Usage: Unlike regular HttpResponse objects in Django, you don't give Response() class rendered content directly. Instead, you provide it with unrendered data, typically Python data types like lists or dictionaries.

• Serialization: The Response() class cannot handle complex data types directly, like Django model instances. You need to convert these complex types into simpler data types (serialization) before passing them to Response().

• Data Serialization: Use Django REST framework's Serializer classes to convert complex data (like Django models) into Python primitives (like dictionaries). This prepares the data for the Response() object to handle.

• Arguments:

• data: Serialized data (Python primitives) that will be sent in the response.
• status: HTTP status code for the response (defaults to 200 for OK). It tells the client whether the request was successful or had an error.
• template_name: Optional HTML template name to use if rendering HTML responses.
• headers: Additional HTTP headers to include in the response.
• content_type: The type of content in the response (usually set automatically based on content negotiation).

Examples

1. Sending JSON Data: ```python from rest_framework.response import Response from rest_framework.decorators import api_view

@api_view(['GET']) def get_books(request): books = [{'title': 'Book 1', 'author': 'Author A'}, {'title': 'Book 2', 'author': 'Author B'}] return Response(books) `` - Here,Response()` is used to send a list of books as JSON data.

1. Handling Error Responses: ```python from rest_framework.response import Response from rest_framework import status

def create_book(request): if request.method == 'POST': # Some logic to create a book if book_created_successfully: return Response({'message': 'Book created successfully'}, status=status.HTTP_201_CREATED) else: return Response({'error': 'Failed to create book'}, status=status.HTTP_400_BAD_REQUEST) `` - In this example,Response()` is used to send messages about the success or failure of creating a book, along with appropriate HTTP status codes.

Notes

• Flexibility: Response() allows your Django API to respond with data in different formats based on client needs.
• Serialization: Use serializers to convert complex data into formats Response() can handle.
• HTTP Status Codes: Always consider setting appropriate HTTP status codes to inform clients about the success or failure of their requests.

Using Response() in Django REST framework simplifies handling API responses, ensuring data is sent back to clients in the right format with proper status information.

Sure, let's break down what that means:

HTTP Content Negotiation: This is the process where a server and a client agree on the format of data that will be exchanged in an HTTP request. It's like deciding on the language in which two people will communicate.

Response Class: In REST framework (used with Django), the Response class helps you send data back to clients in different formats (like JSON, HTML, etc.) based on what the client requests.

Example: Imagine you have an endpoint /api/books/ that lists books. When a client (like a web browser or mobile app) asks for this list, they might want the data in JSON format (which is common for APIs), while another client might prefer HTML (for displaying in a web browser).

Using Response Class: Instead of manually crafting the response each time, you can use the Response class provided by REST framework. It makes it easier to handle different formats. For instance, if a client requests JSON, the Response class can automatically convert your Python data (like lists of books) into JSON format.

Why Use It: By using the Response class, you ensure that your API can easily respond with data in the format that the client prefers, whether it's JSON, HTML, or another format. It simplifies your code and makes your API more flexible for different clients.

When to Use It: Unless you have specific reasons not to, it's recommended to use the Response class in your views that handle API requests. This way, REST framework can handle content negotiation smoothly, ensuring the right format is sent back to the client without you having to handle all the details manually.

In summary, HTTP content negotiation and the Response class in REST framework help you efficiently manage how data is sent and received between your Django application and its clients, ensuring flexibility and ease of use.

I should be interested in how these kinds of articles are written to attract public audiences

Author response:

First we thank the reviewers for a thorough reading of our paper and some useful comments. A recurrent remark of the reviewers concerns the appearance of kRas-expressing cells (labelled by a nuclear blue fluorescent marker) which we attribute to the progeny of the initially induced cell. The reviewers suggest that these cells may have been obtained through activation of the Cre-recombinase in other cells by cyclofen released from light scattering, via diffusion, leakiness, etc. These remarks are perfectly reasonable from people not familiar with the cyclofen uncaging approach that we are using but are unwarranted as we shall show below.

We have been using cyclofen uncaging with subsequent activation of a Cre-recombinase (or some other proteins) since 2010 (see ref.34, Sinha et al., Zebrafish 7, 199-204 (2010) and our 2018 review (ref.35, Zhang et al., ChemBioChem 19,1-8 (2018)). In our experiments, the embryos are incubated in the dark in 6M caged cyclofen (cCyc) and washed in E3 medium (or transferred to a new medium with no cCyc). In these conditions, over many years we never observed activation of the recombinase, i.e. the appearance of the associated fluorescent label in cells of embryos grown in E3 medium. Hence leakiness can be ruled out (in presence of cCyc or in its absence).

Following transfer of the embryos to new E3 medium we illuminate the embryos locally with light at 405nm. In these conditions, cCyc is only partially uncaged and results in activation of Cre-recombinase in only a few cells (1,2, 3, …) within the illuminated region only, namely in the appearance of the kRas-associated nuclear blue fluorescent label in usually one cell (and sometimes in a few more; data and statistics will be incorporated in a revised manuscript). In absence of any further treatment (e.g. activation of a reprogramming factor) these fluorescently labelled cells disappear within a few days (either via shut-down of their promotor, apoptosis or some other mechanism). The crucial point here is that we see less and not more kRas expressing cells (i.e. with nuclear blue fluorescence). This observation rules out activation of Cre-recombinase in other cells days after illumination due to leakiness, cyclofen released by light or diffusing from the illumination spot.

To observe many more fluorescent cells days after activation of the initial cell, one needs to transiently activate VentX-GR by overnight incubation in dexamethasone (DEX) (Injecting the embryos at 1-cell stage with VentX-GR or incubating them in DEX does not result in the appearance of more blue fluorescent cells). Following activation of VentX-GR, the fluorescent cells observed a couple of days after initiation are visualized in E3 medium (i.e. in absence of cyclofen) and are localized to the vicinity of the otic vesicle (the region where the initial cell was activated). In a revised manuscript we will present images of these fluorescent cells taken a few days apart from the same embryo in which a single cell was initially activated. Hence, we attribute these cells to the progeny of the activated cell. Obviously, single cell tracking via time-lapse microscopy would nail down this issue and provide fascinating insight into the initial stages of tumor growth. Unfortunately, immobilization of embryos in the usual medium (e.g. MS222, tricaine) over 5-6 days to track the division and motion of single cells is not possible. We are considering some other possibilities (immobilization in bungarotoxin or via photo-activation of anionic channels), but these challenging experiments are for a future paper.

Reviewer #1 (Public Review):

The authors then performed allotransplantations of allegedly single fluorescent TICs in recipient larvae and found a large number of fluorescent cells in distant locations, claiming that these cells have all originated from the single transplanted TIC and migrated away. The number of fluorescent cells showed in the recipient larve just after two days is not compatible with a normal cell cycle length and more likely represents the progeny of more than one transplanted cell.

As mentioned in the manuscript, we measure the density of cells/nl and inject in the yolk of 2dpf Nacre embryos a volume containing about 1 cell, following published protocols (S.Nicoli and M.Presta, Nat.Prot. 2,2918 (2007)). We further image the injected cell(s) by fluorescence microscopy immediately following injection, as shown in Fig.4A and Fig.S8B. We might miss a few cells but not many. With a typical cell cycle of ~10h the images of tumors in larvae at 3dpt (and not 2dpt as misunderstood by this reviewer) correspond to ~100 cells. In any case the purpose of this experiment was not to study tumorigenesis upon transplantation but to show that the progeny of the initially induced cells is capable of developing into a tumor in a naïve fish, which is the operational definition of cancer that we adopted here.

The ability to migrate from the injection site should be documented by time-lapse microscopy.

As stated above our purpose here is not to study tumor formation from transplanted cell(s) but to use that assay as an operational test of cancer. Besides as mentioned earlier single cell tracking in larvae over 3-4dpt is not a trivial task.

Then, the authors conclude that "By allowing for specific and reproducible single cell malignant transformation in vivo, their optogenetic approach opens the way for a quantitative study of the initial stages of cancer at the single cell level". However, the evidence for these claims are weak and further characterization should be performed to:

(1) show that they are actually activating the oncogene in a single cell (the magnification is too low and it is difficult to distinguish a single nucleus, labelling of the cell membrane may help to demonstrate that they are effectively activating the oncogene in, or transplanting, a single cell)

In a revised manuscript we will provide larger magnification of the initial induced cell and show examples of oncogene activation in more than one cell.

(2) the expression of the genes used as markers of tumorigenesis is performed in whole larvae, with only a few transformed cells in them. Changes should be confirmed in FACS sorted fluorescent cells

When the oncogene is activated in a whole larvae all cells are fluorescent and thus FACS is of no use for cell sorting. Sorting could be done in larvae where single cells are activated, but then the efficiency of FACS is not good enough to isolate the few fluorescent cells among the many more non-fluorescent ones. We agree that the change in expression of the genes used as markers of tumorigenesis is an underestimate of their true change, but our goal at this time is not to precisely measure the change in expression level, but to show that the pattern of change is different from the controls and corresponds to what is expected in tumorigenesis.

(3) the histology of the so called "tumor masses" is not showing malignant transformation, but at the most just hyperplasia.

The histology of the hyperplasic tissues displays cellular proliferation with a higher density of nuclear material which is characteristic of tumors, Fig.S4C. Besides the increased expression of pERK in these tissues, Fig.S4A,B is also a hallmark of cancer.

In the brain, the sections are not perfectly symmetrical and the increase of cellularity on one side of the optic tectum is compatible with this asymmetry.

The expected T-shape formed by the sections of the tegmentum and hypothalamus are compatible with the symmetric sections shown in Fg.2D. The asymmetry in the optic tectum is a result of the hyperplasic growth.

(4) The number of fluorescent cells found dispersed in the larvae transplanted with one single TIC after 48 hours will require a very fast cell cycle to generate over 50 cells. Do we have an idea of the cell cycle features of the transplanted TICs?

As answered above, the transplanted larvae are shown at 3dpt (and not 2dpt as misunderstood by this reviewer). With a cell cycle of about 10h, a single cell can give rise to about 100 cells in that time lapse.

Reviewer #2 (Public Review):

Summary:

This paper describes a genetically tractable and modifiable system …which could be used to study an array of combinations and temporal relationships of these cancer drivers/modifiers.

We thank this referee for its positive comments. We would also like to point out that our approach provides for the first quantitative means to estimate the probability of tumorigenesis from a single cell, an estimate which is crucial in any assessment of cancer malignancy and the effectiveness of prophylactics.

Weaknesses:

There is minimal quantitation of … the efficiency of activation of the Ras-TFP fusion (Fig 1) in, purportedly, a single cell. …, such information seems essential.

In a revised manuscript we will add more images of induction of a single (or a few cells) and a table where the efficiency of RAS activation is detailed.

The authors indicate that a single cell is "initiated" (Fig 2) using the laser optogenetic technique, but without definitive genetic lineage tracing, it is not possible to conclude that cells expressing TFP distant from the target site near the ear are daughter cells of the claimed single "initiated" cell. A plausible alternative explanation is 1) that the optogenetic targeting is more diffuse (i.e. some of the light of the appropriate wavelength hits other cells nearby due to reflection/diffraction), so these adjacent cells are additional independent "initiated" cells or 2) that the uncaged tamoxifen analogue can diffuse to nearby cells and allow for CreER activation and recombination.

We have addressed this point in our general comments to the reviewers’ remarks. The possibilities mentioned by this reviewer would result in cells expressing TFP in absence of VentX activation, which is not the case. Cells expressing TFP away from the initial site are observed days after activation of the oncogene (and TFP) in a single cell and only upon activation of VentX.

In Fig 2B, the claim is made that "the activated cell has divided, giving rise to two cells" - unless continuously imaged or genetically traced, this is unproven.

We have addressed this remark previously. Tracking of larvae over many days is not possible with the usual protocol using tricaine to immobilize the larvae. Nonetheless, in a revised version we will present images of an embryo imaged at various times post activation where proliferation of the cells can be observed. We are pursuing other alternatives for time-lapse microscopy over many days since, besides convincing the sceptics, a single cell tracking experiment (possibly coupled with in-situ spatial transcriptomics) will shed a new and fascinating light on the initial stages of tumor growth.

In addition, it appears that Figures S3 and S4 are showing that hyperplasia can arise in many different tissues (including intestine, pancreas, and liver, S4C) with broad Ras + Ventx activation …. This should be clarified in the manuscript).

This is true and will be clarified in the new version.

In Fig S7 where single cell activation and potential metastasis is discussed, similar gut tissues have TFP+ cells that are called metastatic, but this seems consistent with the possibility that multiple independent sites of initiation are occurring even when focal activation is attempted.

As mentioned previously this is ruled out by the fact that these cells are observed days after cyclofen uncaging (and TFP activation) and if and only if VentX is activated.

Although the hyperplastic cells are transplantable (Fig 4), the use of the term "cells of origin of cancer" or metastatic cells should be viewed with care in the experiments showing TFP+ cells (Fig 1, 2, 3) in embryos with targeted activation for the reasons noted above.

The purpose of this transplantation experiment was to show that cell in which both kRas and VentX have been activated possess the capacity to metastasize and develop a tumor mass when transplanted in a naïve zebrafish. This - to the best of our knowledge - is the operational definition of a malignant tumor.

Reviewer #3 (Public Review):

Summary:

This study employs an optogenetics approach … to examine tumourigenesis probabilities under altered tissue environments.

We thank this reviewer for this remark, since we believe that the opportunity to assess the probability of tumorigenesis from a single cell is possibly the most significant contribution of this work. To the best of our knowledge this has never been done before.

Weaknesses:

Lack of Methodological Clarity: The manuscript lacks detailed descriptions of methodologies,

In a revised manuscript we will include additional detail of our methodology.

Sub-optimal Data Presentation and Quality:

Lack of quantitative data and control condition data obtained from images of higher magnification limits the ability to robustly support the conclusions.

In a revised version we will include more images at higher magnification and quantitative data to support the main report of targeted single cell induction.

Here are some details:

Authors might want to provide more evidence to support their claim on the single cell KRAS activation.

More images and a data on activation of single or few cells in the illumination field will be provided in a revised version.

· Stability of cCYC: The manuscript does not provide information on the half-life and stability of cCYC. Understanding these properties is crucial for evaluating the system's reliability and the likelihood of leakiness, which could significantly influence the study's outcomes.

We have been using the cCyc system for about 14 years. We refer the reader to our previous papers and reviews on this methodology (e.g. ref. 34,35). Briefly, cCyc is stable when not illuminated with light around 375nm. Typically, we incubate our embryos in the dark for about 1h before transferring them into E3 medium and illuminating them. Assessing the leakiness of the system is easy as expression of the fluorescent marker is permanently turned on. We have observed none in the conditions of our experiment.

· Metastatic Dissemination claim: However, the absence of a supportive cellular compartment within the fin-fold tissue makes the presence of mTFP-positive metastatic cells there particularly puzzling. This distribution raises concerns about the spatial specificity of the optogenetic activation protocol … The unexpected locations of these signals suggest potential ectopic activation of the KRAS oncogene,

We have addressed this remark in the introduction and above. Specifically, metastatic and proliferative mTFP-positive cells are observed if and only if VentX is also activated concomitant with activation of kRAS in a single cell. No proliferative cells are observed in absence of VentX activation, or in presence of VentX or Dex alone, or if kRAS has not been activated by cyclofen uncaging.

· Image Resolution Concerns: The cells depicted in Figure 3C β, which appear to be near the surface of the yolk sac and not within the digestive system as suggested in the MS, underscore the necessity for higher-resolution imaging. Without clearer images, it is challenging to ascertain the exact locations and states of these cells, thus complicating the assessment of experimental results.

Better images will be provided in the revised version.

· The cell transplantation experiment is lacking protocol details:

Details will be provided in the revised version. We have followed regular protocols for transplantation: S.Nicoli and M.Presta, Nat.Prot. 2,2918 (2007).

• If the cells are obtained from whole larvae with induced RAS + VX expression, it is notable and somewhat surprising that the larvae survived up to six days post-induction (6dpi) before cells were harvested for transplantation. This survival rate and the subsequent ability to obtain single cell suspensions raise questions about the heterogeneity of the RAS + VX expressing cells that transplanted.

From Fig.S4D, about 50% of the embryos survive at 6dpi. Though an interesting question by itself we have not (yet) addressed the important issue of the heterogeneity of the outgrowth obtained from a single cell. Our purpose here was just to show that cells in which both kRAS and VentX have been activated possess the capacity to metastasize and develop a tumor mass when transplanted in a naïve zebrafish. This - to the best of our knowledge - is the operational definition of a malignant tumor.

· Unclear Experimental Conditions in Figure S3B: …It is not specified whether the activation of KRAS was targeted to specific cells or involved whole-body exposure.

This was whole body (global) illumination and will be specified in the revised version.

· Contrasting Data in Figure S3C compared to literature: The graph in Figure S3C indicates that KRAS or KRAS + DEX induction did not result in any form of hyperplastic growth. The authors should provide detailed descriptions of the conditions under which the experiments were conducted in Figure S3B and clarifying the reasons for the discrepancies observed in Figure S3C are crucial. The authors should discuss potential reasons for the deviation from previous reports.

This discrepancy will be discussed in the revised version. First the previous reports consider the development of tumors over a longer time-span (4-5 weeks) which we have not studied here. Second, the expression of the oncogene in these reports might be stronger than in ours. Third, the stochastic appearance of tumors in these reports suggest that some other mechanism (transient stress-induced reprogramming?) might have activated the oncogene in the initial cell.

Further comments:

Throughout the study, KRAS-activated cell expansion and metastasis are two key phenotypes discussed that Ventx is promoting. However, the authors did not perform any experiments to directly show that KRAS+ cells proliferate only in Ventx-activated conditions.

Yes, we did. See Fig. S1 and compare with Fig.S3B, or Fig.S8A in comparison with Fig.2A,B.

The authors also did not show any morphological features or time-lapse videos demonstrating that KRAS+ cells are motile, even though zebrafish is an excellent model for in vivo live imaging. This seems to be a missed opportunity for providing convincing evidence to support the authors' conclusions.

Performing single cell time-lapse microscopy on larvae over many (4-5) days is not possible with the regular tricaine protocol for immobilization. We are definitely planning such experiments, but they will require some other protocol, perhaps using bungarotoxin or some optogenetic inhibitory channels. Nonetheless, in the revised version we will show images of the same embryos at various times post single cell induction displaying proliferation of cells.

There were minimal experimental details provided for the qPCR data presented in the supplementary figures S5 and S6, therefore, it is hard to evaluate result obtained.

More details will be given in the revised version.

This report provides a detailed list of all transactions (payments, refunds, and adjustments) against every settlement in the selected time range with additional POS details

est

Contudo não podemos esquecer que os objetivos das e-atividades variam de acordo com cada disciplina/módulo. Tendo sempre em conta que devem promover a aprendizagem ativa, desenvolver habilidades especificas, avaliar o conhecimento, estimular a interação de todos num ambiente colaborativo, flexibilizando sempre a aprendizagem, adaptando-a a cada aluno/estilo de vida.

Este parágrafo resume aquilo que tem vindo a ser apresentado sobre a docência digital ao longo deste curso e mudou muito a minha percepção sobre a docência digital. Vai também muito ao encontro daquilo que já há muito tempo defendo para o modo convencional de aprendizagem presencial onde há claramente uma orientação para o modelo expositivo em que o trabalho fica quase exclusivamente do lado do aluno.

Esta ferramenta parece-me muito interessante, e pode ajudar a construir uma e-actividade que responda a estes requisitos aqui sintetizados neste parágrafo.

hustle and bustle - суматоха

文の前半と後半で主語が変わって読みにくいです

代案

ユーザーにファイルが存在しないことを知らせる必要がある場合もあれば

typo どのように

好みの問題化もしれませんが、違和感がありました

代案

可能性が出てきます

頻度が多いほうの「よく書かれ」と見分けがつかないので別の言い回しを考えました

代案

じょうずに書かれ

日本語Windowsでは円マーク（￥）です

ファイルを行単位で処理したい　はいかがでしょうか

異常終了する　はいかがでしょうか

Watch Video/Animation

мы кидаем слово с 21 битом. проводим еще раз алогос и если получаем уже другие контрольные биты. то что-то пошло не так и нужно искать что именно пошло не так. а именно просто суммируем измененные биты. точнее их индексы и получаем индекс, где бит был изменён

то есть мы знаем контрольные биты

Далее находим все биты, который контролирует контрольный бит. а он контролирует каждый N бит интервалом в N. где N - это номер (индекс, но с 0, а 1) самого бита. и суммируем все биты. и если в итоге в сумма чётное число, то ставим 0. иначе 1. здесь просто используется бит чётности

が　でしょうか

この本では変数へのattachは「関連付ける」と訳していると思うので揃えたほうがよいのでは

かなり翻訳調なのが気になりました。

代案

self変数が何を指しているのかという質問をよく見かけます。

workflowの訳語は悩ましいですね。

続く説明の中ではモデル化する方法について書かれているので、「モデル化の方法を見つける」としてはいかがでしょうか

～を～を　が気になりました。

代案

別のクラスを継承するには

注意深く　のほうがしっくりくる気がします

「運転」のほうがしっくりくるかも

ちょっと違和感あり

代案

を活用する

categoryの訳語、悩みますね。

代案（自信ない）

仲間

Highlight the text, then select annotate from the pop-up options. Enter your note, then click Post.

少しこなれていない感じがしたので、以下ではどうでしょう。

変更したい設定を正確に指定するのがとても簡単になります。

↓typo

モジュール

↓typo

テンプレート

「グラフ呼び出しをさらにカスタマイズ」としてはどうでしょう。

トル

「〜的」という表現が2回続いているのが少しこなれていない感じがしました。以下のようにしてはどうでしょう。

視覚的に魅力がある表現

Update v3.1(A)

New Recommendation: * For patients with brain metastases that are treated and stable, referral for lifileucel can be considered. Patients with active brain metastases are ineligible for lifileucel/TIL therapy, unless done as part of a clinical trial.

なんだか読みにくいと感じました。

代案（ちょい意訳）

関数で実現したいことができていれば、それだけで十分です。

説明では「～を返すことができます」と書いているので、ここも「～を返す」にしてはどうでしょうか

※以下同様

これでも動きますが、「イヌ」のほうがいいかも

https://youtu.be/QpedwI4Pf0c

https://youtu.be/L8eihSImkO8

https://youtu.be/zkrXXHcKjzs

eLife assessment

In this study, Tutak and colleagues set out to identify factors that mediate Repeat Associated Non-AUG (RAN) translation of CGG repeats in the FMR1 mRNA which are implicated in toxic protein accumulation that underpins ensuing neurological pathologies. This is a useful article that suggests that RPS26 may be implicated in mediating the RAN translation of FMR1 mRNA. However, the evidence supporting the proposed mechanism is incomplete, since the provided data only partially support the authors' conclusion.

Reviewer #1 (Public Review):

Summary:

In this manuscript, Tutak et al use a combination of pulldowns, analyzed by mass spectrometry, reporter assays, and fluorescence experiments to decipher the mechanism of protein translation in fragile X-related diseases. The topic is interesting and important.

Although a role for Rps26-deficient ribosomes in toxic protein translation is plausible based on already available data, the authors' data are not carefully controlled and thus do not support the conclusions of the paper.

Strengths:

The topic is interesting and important.

Weaknesses:

In particular, there is very little data to support the notion that Rps26-deficient ribosomes are even produced under the circumstances. And no data that indicate that they are involved in the RAN translation. Essential controls (for ribosome numbers) are lacking, no information is presented on the viability of the cells (Rps26 is an essential protein), and the differences in protein levels could well arise from block in protein synthesis, and cell division coupled to differential stability of the proteins.

Specific points:

(1) Analysis of the mass spec data in Supplemental Table S3 indicates that for many of the proteins that are differentially enriched in one sample, a single peptide is identified. So the difference is between 1 peptide and 0. I don't understand how one can do a statistical analysis on that, or how it would give out anything of significance. I certainly do not think it is significant. This is exacerbated by the fact that the contaminants in the assay (keratins) are many, many-fold more abundant, and so are proteins that are known to be mitochondrial or nuclear, and therefore likely not actual targets (e.g. MCCC1, PC, NPM1; this includes many proteins "of significance" in Table S1, including Rrp1B, NAF1, Top1, TCEPB, DHX16, etc...).

The data in Table S6/Figure 3A suffer from the same problem.

I am not convinced that the mass spec data is reliable.

(2) The mass-spec data however claims to identify Rps26 as a factor binding the toxic RNA specifically. The rest of the paper seeks to develop a story of how Rps26-deficient ribosomes play a role in the translation of this RNA. I do not consider that this makes sense.

(3) Rps26 is an essential gene, I am sure the same is true for DHX15. What happens to cell viability? Protein synthesis? The yeast experiments were carefully carried out under experiments where Rps26 was reduced, not fully depleted to give small growth defects.

(4) Knockdown efficiency for all tested genes must be shown to evaluate knockdown efficiency.

(5) The data in Figure 1E have just one mock control, but two cell types (control si and Rps26 depletion).

(6) The authors' data indicate that the effects are not specific to Rps26 but indeed also observed upon Rps25 knockdown. This suggests strongly that the effects are from reduced ribosome content or blocked protein synthesis. Additional controls should deplete a core RP to ascertain this conclusion.

(7) Supplemental Figure S3 demonstrates that the depletion of S26 does not affect the selection of the start codon context. Any other claim must be deleted. All the 5'-UTR logos are essentially identical, indicating that "picking" happens by abundance (background).

(8) Mechanism is lacking entirely. There are many ways in which ribosomes could have mRNA-specific effects. The authors tried to find an effect from the Kozak sequence, unsuccessfully (however, they also did not do the experiment correctly, as they failed to recognize that the Kozak sequence differs between yeast, where it is A-rich, and mammalian cells, where it is GGCGCC). Collisions could be another mechanism.

Reviewer #2 (Public Review):

Summary:

Translation of CGG repeats leads to the accumulation of poly G, which is associated with neurological disorders. This is a valuable paper in which the authors sought out proteins that modulate RAN translation. They determined which proteins in Hela cells bound to CGG repeats and affected levels of polyG encoded in the 5'UTR of the FMR1 mRNA. They then showed that siRNA depletion of ribosomal protein RPS26 results in less production of FMR1polyG than in control. There are data supporting the claim that RPS26 depletion modulates RAN translation in this RNA, although for some results, the Western results are not strong. The data to support increased aggregation by polyG expression upon S26 KD are incomplete.

Strengths:

The authors have proteomics data that show the enrichment of a set of proteins on FMR1 RNA but not a related RNA.

Weaknesses:

-It is insinuated that RPS26 binds the RNA to enhance CGG-containing protein expression. However, RPS26 reduction was also shown previously to affect ribosome levels, and reduced ribosome levels can result in ribosomes translating very different RNA pools.

-A significant claim is that RPS26 KD alleviates the effects of FMR polyG expression, but those data aren't presented well.

Reviewer #3 (Public Review):

Tutak et al provide interesting data showing that RPS26 and relevant proteins such as TSR2 and RPS25 affect RAN translation from CGG repeat RNA in fragile X-associated conditions. They identified RPS26 as a potential regulator of RAN translation by RNA-tagging system and mass spectrometry-based screening for proteins binding to CGG repeat RNA and confirmed its regulatory effects on RAN translation by siRNA-based knockdown experiments in multiple cellular disease models and patient-derived fibroblasts. Quantitative mass spectrometry analysis found that the expressions of some ribosomal proteins are sensitive to RPS26 depletion while approximately 80% of proteins including FMRP were not influenced. Since the roles of ribosomal proteins in RAN translation regulation have not been fully examined, this study provides novel insights into this research field. However, some data presented in this manuscript are limited and preliminary, and their conclusions are not fully supported.

(1) While the authors emphasized the importance of ribosomal composition for RAN translation regulation in the title and the article body, the association between RAN translation and ribosomal composition is apparently not evaluated in this work. They found that specific ribosomal proteins (RPS26 and RPS25) can have regulatory effects on RAN translation（Figures 1C, 2B, 2C, 2E, 4A, 5A, and 5B), and that the expression levels of some ribosomal proteins can be changed by RPS26 knockdown (Figure 3B, however, the change of the ribosome compositions involved in the actual translation has not been elucidated). Therefore, their conclusive statement, that is, "ribosome composition affects RAN translation" is not fully supported by the presented data and is misleading.

(2) The study provides insufficient data on the mechanisms of how RPS26 regulates RAN translation. Although authors speculate that RPS26 may affect initiation codon fidelity and regulate RAN translation in a CGG repeat sequence-independent manner (Page 9 and Page 11), what they really have shown is just identification of this protein by the screening for proteins binding to CGG repeat RNA (Figure 1A, 1B), and effects of this protein on CGG repeat-RAN translation. It is essential to clarify whether the regulatory effect of RPS26 on RAN translation is dependent on CGG repeat sequence or near-cognate initiation codons like ACG and GUG in the 5' upstream sequence of the repeat. It would be better to validate the effects of RPS26 on translation from control constructs, such as one composed of the 5' upstream sequence of FMR1 with no CGG repeat, and one with an ATG substitution in the 5' upstream sequence of FMR1 instead of near-cognate initiation codons.

(3) The regulatory effects of RPS26 and other molecules on RAN translation have all been investigated as effects on the expression levels of FMRpolyG-GFP proteins in cellular models expressing CGG repeat sequences （Figures 1C, 2B, 2C, 2E, 4A, 5A, and 5B). In these cellular experiments, there are multiple confounding factors affecting the expression levels of FMRpolyG-GFP proteins other than RAN translation, including template RNA expression, template RNA distribution, and FMRpolyG-GFP protein degradation. Although authors evaluated the effect on the expression levels of template CGG repeat RNA, it would be better to confirm the effect of these regulators on RAN translation by other experiments such as in vitro translation assay that can directly evaluate RAN translation.

(4) While the authors state that RPS26 modulated the FMRpolyG-mediated toxicity, they presented limited data on apoptotic markers, not cellular viability (Figure 1E), not fully supporting this conclusion. Since previous work showed that FMRpolyG protein reduces cellular viability (Hoem G et al., Front Genet 2019), additional evaluations for cellular viability would strengthen this conclusion.

Author response:

Public Reviews:

Reviewer #1 (Public Review):

Summary:

In this manuscript, Tutak et al use a combination of pulldowns, analyzed by mass spectrometry, reporter assays, and fluorescence experiments to decipher the mechanism of protein translation in fragile X-related diseases. The topic is interesting and important.

Although a role for Rps26-deficient ribosomes in toxic protein translation is plausible based on already available data, the authors' data are not carefully controlled and thus do not support the conclusions of the paper.

Strengths:

The topic is interesting and important.

Weaknesses:

In particular, there is very little data to support the notion that Rps26-deficient ribosomes are even produced under the circumstances. And no data that indicate that they are involved in the RAN translation. Essential controls (for ribosome numbers) are lacking, no information is presented on the viability of the cells (Rps26 is an essential protein), and the differences in protein levels could well arise from block in protein synthesis, and cell division coupled to differential stability of the proteins.

We agree that presented data could benefit from addition of suggested experiments. We will  address the ribosome content, global translation rate and cell viability upon RPS26 depletion. We are also planning to apply polysome profiling to determine if RPS26-depleted ribosomes are translationally active.

Specific points:

(1) Analysis of the mass spec data in Supplemental Table S3 indicates that for many of the proteins that are differentially enriched in one sample, a single peptide is identified. So the difference is between 1 peptide and 0. I don't understand how one can do a statistical analysis on that, or how it would give out anything of significance. I certainly do not think it is significant. This is exacerbated by the fact that the contaminants in the assay (keratins) are many, many-fold more abundant, and so are proteins that are known to be mitochondrial or nuclear, and therefore likely not actual targets (e.g. MCCC1, PC, NPM1; this includes many proteins "of significance" in Table S1, including Rrp1B, NAF1, Top1, TCEPB, DHX16, etc...).

The data in Table S6/Figure 3A suffer from the same problem.

Tables S3 and S6 show the mass spectrometry output data from MaxQuant analysis  without any flittering.  Certain identifications, i.e. those denoted as contaminants (such as keratins) were removed during statistical analysis in Perseus software. Regarding the data presented in Table S6 (SILAC data), we argue that these data are of very good quality. More than 2000 proteins were identified in a 125min gradient, with over 80% of proteins that were identified with at least 2 unique peptides. However, we acknowledge that the description of Tables S3 and S6 may lead to misunderstanding, thus we will clarify their explanation.

I am not convinced that the mass spec data is reliable.

(2) The mass-spec data however claims to identify Rps26 as a factor binding the toxic RNA specifically. The rest of the paper seeks to develop a story of how Rps26-deficient ribosomes play a role in the translation of this RNA. I do not consider that this makes sense.

Indeed, we identified RPS26 as a protein co-precipitated with FMR1 RNA containing expanded CGG repeats. However, we do not claim that they interact directly. Downregulation of FMRpolyG biosynthesis could be an outcome of the alteration of ribosomal assembly, changes in efficiency and fidelity of PIC scanning or impeded elongation or more likely combination of some of these processes. We will  provide better explanation regarding those issues in the revised version of the manuscript.

(3) Rps26 is an essential gene, I am sure the same is true for DHX15. What happens to cell viability? Protein synthesis? The yeast experiments were carefully carried out under experiments where Rps26 was reduced, not fully depleted to give small growth defects.

We agree with the Reviewer 1 that RPS26 is an essential protein. Previously, it was shown that cell viability in cells with mutated C-terminal deletion of RPS26 is decreased (Havkin-Solomon T, Nucleic Acids Res 2023). We will address the question regarding the suppression of FMRpolyG in models with partial RPS26 knock-down.

(4) Knockdown efficiency for all tested genes must be shown to evaluate knockdown efficiency.

Missing experiments showing efficiency of knock-down will be included in the revised version of the manuscript.

(5) The data in Figure 1E have just one mock control, but two cell types (control si and Rps26 depletion).

We will clarify this ambiguity in the revised version of the manuscripts.

(6) The authors' data indicate that the effects are not specific to Rps26 but indeed also observed upon Rps25 knockdown. This suggests strongly that the effects are from reduced ribosome content or blocked protein synthesis. Additional controls should deplete a core RP to ascertain this conclusion.

We agree that observed effect may stem partially from reduced ribosome content, however, we argue that this is not the only explanation. In the publication concerning RPS25 regulation of G4C2-related RAN translation (Yamada SB, 2019, Nat Neurosci), it was shown that RPS25 KO does not affect global translation. Our experiments (SUnSET assay, unpublished) indicated that RPS26 KD also did not reduce global translation rate significantly. We will present that data in the revised version of the manuscript.

(7) Supplemental Figure S3 demonstrates that the depletion of S26 does not affect the selection of the start codon context. Any other claim must be deleted. All the 5'-UTR logos are essentially identical, indicating that "picking" happens by abundance (background).

Results shown in Fig.S3 does not imply that RPS26 does not affect the selection of start codon context entirely. We just tested a few hypotheses. We decided to test -4 position, because this position was indicated as the most sensitive to RPS26 regulation in yeast (Ferretti M, 2017, Nat Struct Mol Biol). Regarding WebLOGO analysis; we wrote in the manuscript that we did not identify any specific motif or enrichment within analysed transcripts in comparison to background. We will clarify this ambiguity in revised version of the manuscript.

(8) Mechanism is lacking entirely. There are many ways in which ribosomes could have mRNA-specific effects. The authors tried to find an effect from the Kozak sequence, unsuccessfully (however, they also did not do the experiment correctly, as they failed to recognize that the Kozak sequence differs between yeast, where it is A-rich, and mammalian cells, where it is GGCGCC). Collisions could be another mechanism.

As in (7).

Reviewer #2 (Public Review):

Summary:

Translation of CGG repeats leads to the accumulation of poly G, which is associated with neurological disorders. This is a valuable paper in which the authors sought out proteins that modulate RAN translation. They determined which proteins in Hela cells bound to CGG repeats and affected levels of polyG encoded in the 5'UTR of the FMR1 mRNA. They then showed that siRNA depletion of ribosomal protein RPS26 results in less production of FMR1polyG than in control. There are data supporting the claim that RPS26 depletion modulates RAN translation in this RNA, although for some results, the Western results are not strong. The data to support increased aggregation by polyG expression upon S26 KD are incomplete.

Strengths:

The authors have proteomics data that show the enrichment of a set of proteins on FMR1 RNA but not a related RNA.

Weaknesses:

- It is insinuated that RPS26 binds the RNA to enhance CGG-containing protein expression. However, RPS26 reduction was also shown previously to affect ribosome levels, and reduced ribosome levels can result in ribosomes translating very different RNA pools.

We agree that presented data could benefit from addition of some experiments. Therefore we will address questions regarding the ribosome content, global translation rate and cell viability upon RPS26 depletion. We are also planning to apply polysome profiling to determine if RPS26-depleted ribosomes are translationally active. However, we did not state that RPS26 binds directly to RNA with expanded CGG repeats and that this interaction is crucial for translation regulation of studied RNA. We just tested such hypotheses. We will improve the text narration in revised version of the manuscript to make major conclusions clearer.

- A significant claim is that RPS26 KD alleviates the effects of FMRpolyG expression, but those data aren't presented well.

We thank the Reviewer 2 for this comment. We will show the data derived from a few different cell models that we already have obtained. Moreover, we will include results of experiments with luminescence readout for FMRpolyG fused with luciferase upon RPS26 KD.

Reviewer #3 (Public Review):

Tutak et al provide interesting data showing that RPS26 and relevant proteins such as TSR2 and RPS25 affect RAN translation from CGG repeat RNA in fragile X-associated conditions. They identified RPS26 as a potential regulator of RAN translation by RNA-tagging system and mass spectrometry-based screening for proteins binding to CGG repeat RNA and confirmed its regulatory effects on RAN translation by siRNA-based knockdown experiments in multiple cellular disease models and patient-derived fibroblasts. Quantitative mass spectrometry analysis found that the expressions of some ribosomal proteins are sensitive to RPS26 depletion while approximately 80% of proteins including FMRP were not influenced. Since the roles of ribosomal proteins in RAN translation regulation have not been fully examined, this study provides novel insights into this research field. However, some data presented in this manuscript are limited and preliminary, and their conclusions are not fully supported.

(1) While the authors emphasized the importance of ribosomal composition for RAN translation regulation in the title and the article body, the association between RAN translation and ribosomal composition is apparently not evaluated in this work. They found that specific ribosomal proteins (RPS26 and RPS25) can have regulatory effects on RAN translation（Figures 1C, 2B, 2C, 2E, 4A, 5A, and 5B), and that the expression levels of some ribosomal proteins can be changed by RPS26 knockdown (Figure 3B, however, the change of the ribosome compositions involved in the actual translation has not been elucidated). Therefore, their conclusive statement, that is, "ribosome composition affects RAN translation" is not fully supported by the presented data and is misleading.

We thank Reviewer 3 for critical comments and suggestions. We agree that the proposed title may be misleading and the presented data does not fully support the aforementioned statement regarding ribosomal composition affecting FMRpolyG synthesis. Hence, we will change the title together with a narrative regarding these unfortunate statements that go beyond the presented results.

(2) The study provides insufficient data on the mechanisms of how RPS26 regulates RAN translation. Although authors speculate that RPS26 may affect initiation codon fidelity and regulate RAN translation in a CGG repeat sequence-independent manner (Page 9 and Page 11), what they really have shown is just identification of this protein by the screening for proteins binding to CGG repeat RNA (Figure 1A, 1B), and effects of this protein on CGG repeat-RAN translation. It is essential to clarify whether the regulatory effect of RPS26 on RAN translation is dependent on CGG repeat sequence or near-cognate initiation codons like ACG and GUG in the 5' upstream sequence of the repeat. It would be better to validate the effects of RPS26 on translation from control constructs, such as one composed of the 5' upstream sequence of FMR1 with no CGG repeat, and one with an ATG substitution in the 5' upstream sequence of FMR1 instead of near-cognate initiation codons.

We will address the question regarding the influence of the content of CGG repeats and START codon selection (including different near-cognate start codons) on RPS26-sensitive translation, and present these data in revised version of the manuscript.

(3) The regulatory effects of RPS26 and other molecules on RAN translation have all been investigated as effects on the expression levels of FMRpolyG-GFP proteins in cellular models expressing CGG repeat sequences Figures 1C, 2B, 2C, 2E, 4A, 5A, and 5B). In these cellular experiments, there are multiple confounding factors affecting the expression levels of FMRpolyG-GFP proteins other than RAN translation, including template RNA expression, template RNA distribution, and FMRpolyG-GFP protein degradation. Although authors evaluated the effect on the expression levels of template CGG repeat RNA, it would be better to confirm the effect of these regulators on RAN translation by other experiments such as in vitro translation assay that can directly evaluate RAN translation.

We agree that there are multiple factors affecting final translation of investigated mRNA including aforementioned processes. We evaluated the level of FMR1 mRNA, which turned out not to be affected upon RPS26 depletion (Figure 2B&C), however, we will address other possibilities as well.

(4) While the authors state that RPS26 modulated the FMRpolyG-mediated toxicity, they presented limited data on apoptotic markers, not cellular viability (Figure 1E), not fully supporting this conclusion. Since previous work showed that FMRpolyG protein reduces cellular viability (Hoem G et al., Front Genet 2019), additional evaluations for cellular viability would strengthen this conclusion.

We thank Reviewer 3 for this suggestion. We addressed the effect of RPS26 KD on apoptotic process induced by FMRpolyG. We will perform other experiments regarding different aspects of FMRpolyG-mediated cell toxicity as well.

Trade-offs Between Views and ViewSets

When deciding whether to use views or ViewSets in Django REST framework, it's important to understand the benefits and drawbacks of each approach. Both have their own use cases and can be more suitable in different scenarios.

Views

Pros: 1. Explicit and Customizable: You have full control over each view. This allows you to handle complex logic and special cases more easily. 2. Fine-grained Control: Allows you to define exactly what each view does, making it easier to optimize performance and security for specific endpoints. 3. Simplicity: For small projects or APIs with a limited number of endpoints, views might be simpler to implement and understand.

Cons: 1. Repetitive Code: You might end up writing a lot of boilerplate code, especially if your API has many endpoints with similar logic. 2. Inconsistent URL Patterns: It's easier to accidentally create inconsistencies in your API's URL patterns and behavior if you're manually defining each endpoint. 3. Maintenance: As your project grows, maintaining a large number of individual views can become cumbersome and error-prone.

ViewSets

Pros: 1. Consistency: Ensures that URL patterns and behavior are consistent across your API, following standard REST conventions. 2. Less Boilerplate: Reduces the amount of code you need to write. Common CRUD operations are automatically handled. 3. Easier Refactoring: Grouping related views into a single ViewSet can make it easier to refactor and maintain your code. 4. DRY Principle: Helps to keep your code DRY (Don't Repeat Yourself), reducing redundancy.

Cons: 1. Less Explicit: Abstracts away some of the details, which can make the behavior of your API less explicit and harder to understand at a glance. 2. Customization: While ViewSets are great for standard CRUD operations, they can be less flexible for endpoints that require complex or custom behavior. 3. Learning Curve: For developers new to Django REST framework, understanding the additional layer of abstraction might take some time.

When to Use Views vs. ViewSets

• Use Views When:
• You need fine-grained control over each endpoint.
• Your API has complex or custom behavior that doesn't fit the standard CRUD operations.
• You're building a small project with only a few endpoints.

• You want the explicitness and clarity of defining each view individually.

• Use ViewSets When:

• You want to minimize boilerplate code for standard CRUD operations.
• Consistency across your API is a priority.
• Your API has many endpoints that follow standard REST conventions.
• You prefer to focus on the high-level design of your API rather than the specifics of URL configuration.

Example Scenario

Views Approach: If you're building a small API with a few endpoints that require complex custom behavior, you might choose to define each view individually. This approach gives you full control over each endpoint and makes the logic explicit.

```python class SnippetList(APIView): def get(self, request, format=None): snippets = Snippet.objects.all() serializer = SnippetSerializer(snippets, many=True) return Response(serializer.data)

def post(self, request, format=None):
    serializer = SnippetSerializer(data=request.data)
    if serializer.is_valid():
        serializer.save(owner=request.user)
        return Response(serializer.data, status=status.HTTP_201_CREATED)
    return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)

```

ViewSets Approach: For a larger API with many standard CRUD operations, using ViewSets can save a lot of time and reduce boilerplate code. It ensures that your API follows consistent URL patterns and behavior.

```python class SnippetViewSet(viewsets.ModelViewSet): queryset = Snippet.objects.all() serializer_class = SnippetSerializer permission_classes = [permissions.IsAuthenticatedOrReadOnly, IsOwnerOrReadOnly]

@action(detail=True, renderer_classes=[renderers.StaticHTMLRenderer])
def highlight(self, request, *args, **kwargs):
    snippet = self.get_object()
    return Response(snippet.highlighted)

```

Router Configuration: Using a router simplifies URL configuration, reducing the risk of inconsistencies and making it easier to manage a large number of endpoints.

```python router = DefaultRouter() router.register(r'snippets', SnippetViewSet) router.register(r'users', UserViewSet)

urlpatterns = [ path('', include(router.urls)), ] ```

By considering these trade-offs, you can choose the approach that best fits the needs of your project and team.

Let's break down how to refactor our views using ViewSets and Routers and ensure that everything is wired correctly.

Step-by-Step Refactoring

1. Create ViewSets: Define the UserViewSet and SnippetViewSet.
2. Use @action decorator: Add custom actions to the SnippetViewSet.
3. Bind ViewSets to URLs: Use DefaultRouter to automatically generate URL patterns.

Updated Code

1. Creating the ViewSets

In snippets/views.py:

```python from rest_framework import viewsets, permissions, renderers from rest_framework.decorators import action from rest_framework.response import Response from .models import Snippet from .serializers import SnippetSerializer, UserSerializer from django.contrib.auth.models import User from .permissions import IsOwnerOrReadOnly

class UserViewSet(viewsets.ReadOnlyModelViewSet): """ This viewset automatically provides list and retrieve actions. """ queryset = User.objects.all() serializer_class = UserSerializer

class SnippetViewSet(viewsets.ModelViewSet): """ This viewset automatically provides list, create, retrieve, update, and destroy actions. Additionally, we provide a custom highlight action. """ queryset = Snippet.objects.all() serializer_class = SnippetSerializer permission_classes = [permissions.IsAuthenticatedOrReadOnly, IsOwnerOrReadOnly]

@action(detail=True, renderer_classes=[renderers.StaticHTMLRenderer])
def highlight(self, request, *args, **kwargs):
    snippet = self.get_object()
    return Response(snippet.highlighted)

def perform_create(self, serializer):
    serializer.save(owner=self.request.user)

```

2. Binding ViewSets to URLs Explicitly (if not using a Router)

In snippets/urls.py, you can manually bind the ViewSet actions to URL patterns. This is not necessary if you use a Router but is useful for understanding how things work under the hood:

```python from django.urls import path from rest_framework.urlpatterns import format_suffix_patterns from snippets.views import SnippetViewSet, UserViewSet, api_root

snippet_list = SnippetViewSet.as_view({ 'get': 'list', 'post': 'create' }) snippet_detail = SnippetViewSet.as_view({ 'get': 'retrieve', 'put': 'update', 'patch': 'partial_update', 'delete': 'destroy' }) snippet_highlight = SnippetViewSet.as_view({ 'get': 'highlight' }, renderer_classes=[renderers.StaticHTMLRenderer]) user_list = UserViewSet.as_view({ 'get': 'list' }) user_detail = UserViewSet.as_view({ 'get': 'retrieve' })

urlpatterns = format_suffix_patterns([ path('', api_root), path('snippets/', snippet_list, name='snippet-list'), path('snippets/<int:pk>/', snippet_detail, name='snippet-detail'), path('snippets/<int:pk>/highlight/', snippet_highlight, name='snippet-highlight'), path('users/', user_list, name='user-list'), path('users/<int:pk>/', user_detail, name='user-detail') ]) ```

3. Using Routers

To simplify the URL configuration, use a Router. In snippets/urls.py:

```python from django.urls import path, include from rest_framework.routers import DefaultRouter from snippets import views

Create a router and register our viewsets with it.

router = DefaultRouter() router.register(r'snippets', views.SnippetViewSet, basename='snippet') router.register(r'users', views.UserViewSet, basename='user')

The API URLs are now determined automatically by the router.

urlpatterns = [ path('', include(router.urls)), path('api-auth/', include('rest_framework.urls', namespace='rest_framework')) ] ```

How It Works

1. ViewSets: Group related view logic (list, retrieve, create, update, delete) into a single class.
2. Routers: Automatically generate URL patterns for ViewSets based on common conventions.
3. Custom Actions: Use the @action decorator to add custom endpoints that don't fit the standard CRUD operations.

Benefits

• Simplifies URL Configuration: Routers handle the URL routing automatically.
• Combines Related Views: ViewSets group related views into a single class, reducing redundancy.
• Flexible: Easily add custom actions to ViewSets.

By refactoring to use ViewSets and Routers, your code becomes cleaner, more maintainable, and easier to understand. The framework handles much of the repetitive boilerplate code, allowing you to focus on the unique aspects of your application.

Let's continue with refactoring our views by combining the SnippetList, SnippetDetail, and SnippetHighlight classes into a single SnippetViewSet class. This will make our code more concise and maintainable.

Refactoring Steps

1. Refactor User Views: Combine UserList and UserDetail into UserViewSet.
2. Refactor Snippet Views: Combine SnippetList, SnippetDetail, and SnippetHighlight into SnippetViewSet.
3. Update URLs: Use a router to generate URL patterns automatically.

Step 1: Refactor User Views

In snippets/views.py, refactor the user views:

```python from rest_framework import viewsets from django.contrib.auth.models import User from .serializers import UserSerializer

class UserViewSet(viewsets.ReadOnlyModelViewSet): """ This viewset automatically provides list and retrieve actions. """ queryset = User.objects.all() serializer_class = UserSerializer ```

Step 2: Refactor Snippet Views

In snippets/views.py, refactor the snippet views:

```python from rest_framework import viewsets, renderers from .models import Snippet from .serializers import SnippetSerializer

class SnippetViewSet(viewsets.ModelViewSet): """ This viewset automatically provides list, create, retrieve, update, and destroy actions. Additionally, we also provide an extra highlight action. """ queryset = Snippet.objects.all() serializer_class = SnippetSerializer

@action(detail=True, renderer_classes=[renderers.StaticHTMLRenderer])
def highlight(self, request, *args, **kwargs):
    snippet = self.get_object()
    return Response(snippet.highlighted)

```

Step 3: Update URLs

In snippets/urls.py, register the viewsets with a router and update the URL patterns:

```python from django.urls import path, include from rest_framework.routers import DefaultRouter from snippets import views

Create a router and register our viewsets with it.

router = DefaultRouter() router.register(r'snippets', views.SnippetViewSet) router.register(r'users', views.UserViewSet)

The API URLs are now determined automatically by the router.

Additionally, we include the login URLs for the browsable API.

urlpatterns = [ path('', include(router.urls)), path('api-auth/', include('rest_framework.urls', namespace='rest_framework')) ] ```

How It Works

1. UserViewSet: Combines the UserList and UserDetail views into a single viewset that handles read-only operations.
2. SnippetViewSet: Combines the SnippetList, SnippetDetail, and SnippetHighlight views into a single viewset. The highlight action is defined as a custom action within the viewset.
3. Router: The DefaultRouter generates URL patterns automatically, so you don't have to define them manually.

Example in Action

1. User Request:
2. GET /users/: Lists all users (handled by UserViewSet).
3. GET /users/1/: Retrieves details for user with ID 1 (handled by UserViewSet).
4. GET /snippets/: Lists all snippets (handled by SnippetViewSet).
5. GET /snippets/1/: Retrieves details for snippet with ID 1 (handled by SnippetViewSet).
6. GET /snippets/1/highlight/: Retrieves highlighted HTML for snippet with ID 1 (handled by SnippetViewSet).

By refactoring to use ViewSets and Routers, our code becomes cleaner and more maintainable, and we let the framework handle much of the repetitive boilerplate code for us.

Let's break down how to use ViewSets and Routers in Django REST framework to simplify API development.

What are ViewSets and Routers?

• ViewSets: A ViewSet is a class that combines the logic for multiple related views. Instead of defining separate views for listing, retrieving, creating, updating, and deleting objects, you define a single ViewSet class that handles all these actions.
• Routers: A Router automatically generates the URL patterns for the ViewSet, eliminating the need to define the URL patterns manually.

Benefits

• Simplifies URL configuration: Routers handle the URL routing automatically.
• Combines related views: ViewSets group related views into a single class.

Steps to Use ViewSets and Routers

1. Create a ViewSet: Define a ViewSet class that handles the logic for multiple views.
2. Register the ViewSet with a Router: Use a Router to automatically generate URL patterns for the ViewSet.
3. Include the Router's URLs in your project.

Example

Step 1: Create a ViewSet

In snippets/views.py, create a ViewSet for Snippet and User:

```python from rest_framework import viewsets from .models import Snippet from .serializers import SnippetSerializer, UserSerializer from django.contrib.auth.models import User

class SnippetViewSet(viewsets.ModelViewSet): queryset = Snippet.objects.all() serializer_class = SnippetSerializer

class UserViewSet(viewsets.ReadOnlyModelViewSet): queryset = User.objects.all() serializer_class = UserSerializer ```

Explanation: - SnippetViewSet: Handles all CRUD operations for Snippet objects. - UserViewSet: Handles read-only operations for User objects.

Step 2: Register the ViewSet with a Router

In snippets/urls.py, register the ViewSets with a Router:

```python from django.urls import path, include from rest_framework.routers import DefaultRouter from snippets import views

Create a router and register our viewsets with it.

router = DefaultRouter() router.register(r'snippets', views.SnippetViewSet) router.register(r'users', views.UserViewSet)

The API URLs are now determined automatically by the router.

Additionally, we include the login URLs for the browsable API.

urlpatterns = [ path('', include(router.urls)), path('api-auth/', include('rest_framework.urls', namespace='rest_framework')) ] ```

Explanation: - DefaultRouter: Generates URL patterns for the registered ViewSets. - router.register: Registers the SnippetViewSet and UserViewSet with the router. - path('', include(router.urls)): Includes the router-generated URL patterns in the project's URLs. - path('api-auth/', include('rest_framework.urls', namespace='rest_framework')): Adds login URLs for the browsable API.

How It Works

1. ViewSets: Combine the logic for listing, retrieving, creating, updating, and deleting objects into a single class.
2. Routers: Automatically generate the URL patterns for these operations based on common conventions.

Example in Action

1. Request: When a user sends a GET request to /snippets/, the SnippetViewSet handles it and returns a list of snippets.
2. Router: The DefaultRouter automatically routes this request to the correct method in SnippetViewSet.

By using ViewSets and Routers, we simplify the process of creating and managing API endpoints, focusing on the logic rather than URL configurations.

eLife assessment

This fundamental work has completed our understanding of the singular binding profile of the Rhino HP1 protein to chromatin, a key step in converting certain genomic regions into piRNA source loci. The evidence supporting the conclusions is compelling. Phylogenetic analyses, structure prediction, rigorous biochemical assays and in vivo genetics emphasize the importance of the Rhino chromodomain in the recognition of both a histone mark and a DNA-binding protein, and highlight the importance of a single chromodomain residue in the protein-protein interaction.

Reviewer #1 (Public Review):

Summary:

The manuscript focuses on an unexpected finding that a tiny change in a protein's aminoacid sequence can redefine its biological function. The authors' data and analyses explain how a chromodomain, typically implicated in interactions with histones, can also mediate binding of HP1 homolog Rhino to the non-histone partner protein Kipferl. They elegantly pinpoint the capacity for such interaction to a single aminoacid substitution (in fact, a single-nucleotide! substitution).

Strengths:

Both genetic and biochemical approaches are applied to rigorously probe the proposed explanation. The authors find their predictions to be borne out both in vivo, in mutant animals, and in biochemical experiments. The manuscript also features phylogenetic comparisons that put the finding into a broader evolutionary perspective.

Weaknesses pointed out in the original submission were addressed in the revised manuscript.

Reviewer #3 (Public Review):

Summary:

This article is a direct follow-up to the paper published last year in eLife by the same group. In the previous article, the authors discovered a zinc finger protein, Kipferl, capable of guiding the HP1 protein Rhino towards certain genomic regions enriched in GRGGN motifs and packaged in heterochromatin marked by H3K9me3. Unlike other HP1 proteins, Rhino recruitment activates the transcription of heterochromatic regions, which are then converted into piRNA source loci. The molecular mechanism by which Kipferl interacts specifically with Rhino (via its chromodomain) and not with other HP1 proteins remained enigmatic.

In this latest article, the authors go a step further by elucidating the molecular mechanisms important for the specific interaction of Rhino and not other HP1 proteins with Kipferl. A phylogenetic study carried out between the HP1 proteins of 5 Drosophila species led them to study the importance of an AA Glycine at position 31 located in the Rhino chromodomain, an AA different from the AA (aspartic acid) found at the same position in the other HP1 proteins. The authors then demonstrate, through a series of structure predictions, biochemical and genetic experiments, that this specific AA in the Rhino-specific chromodomain explains the difference in the chromatin binding pattern between Rhino and the other Drosophila HP1 proteins. Importantly, the G31D conversion of the Rhino protein prevents interaction between Rhino and Kipferl, phenocopying a Kipfer mutant.

Strengths:

The strength of this study is to test at the molecular and genetic level whether the difference in the AA sequence- encovered by phylogenetic analysis of HP1 proteins including Rhino combined with structure prediction- can explain the difference in chromatin binding patterns between HP1 proteins and Rhino.<br /> To do so they have created a Rhino mutant by introducing a point mutation into the endogenous rhino gene, reverting the Glycine in position 31 to the aspartic acid found in all other HP1 proteins. Even if the Rhino G31D mutant retains its ability to interact with H3K9me3 (predictive and biochemistry approaches that I'm less familiar with) it does not localize correctly on the chromatin preventing certain regions such as locus 80F from being converted into piRNA source loci. However other regions such as satellite regions attract the Rhino mutant protein converting them into super piRNA source loci, phenocopying the effects observed in a Kipferl mutant. Why Rhino when not bound to Kipferl concentrates in satellite regions is a question that remains unanswered.

Weaknesses:

In this new version of the manuscript, the authors have answered all the questions and weaknesses raised previously.

Author response:

The following is the authors’ response to the original reviews.

Public Review:

This article is a direct follow-up to the paper published last year in eLife by the same group. In the previous article, the authors discovered a zinc finger protein, Kipferl, capable of guiding the HP1 protein Rhino towards certain genomic regions enriched in GRGGN motifs and packaged in heterochromatin marked by H3K9me3. Unlike other HP1 proteins, Rhino recruitment activates the transcription of heterochromatic regions, which are then converted into piRNA source loci. The molecular mechanism by which Kipferl interacts specifically with Rhino (via its chromodomain) and not with other HP1 proteins remained enigmatic. 

In this latest article, the authors go a step further by elucidating the molecular mechanisms important for the specific interaction of Rhino and not other HP1 proteins with Kipferl. A phylogenetic study carried out between the HP1 proteins of 5 Drosophila species led them to study the importance of an AA Glycine at position 31 located in the Rhino chromodomain, an AA different from the AA (aspartic acid) found at the same position in the other HP1 proteins. The authors then demonstrate, through a series of structure predictions, biochemical, and genetic experiments, that this specific AA in the Rhino-specific chromodomain explains the difference in the chromatin binding pattern between Rhino and the other Drosophila HP1 proteins. Importantly, the G31D conversion of the Rhino protein prevents interaction between Rhino and Kipferl, phenocopying a Kipferl mutant. 

Strengths: 

The authors' effective use of phylogenetic analyses and protein structure predictions to identify a substitution in the chromodomain that allows Rhino's specific interaction with Kipferl is very elegant. Both genetic and biochemical approaches are applied to rigorously probe the proposed explanation. They used a point mutation in the endogenous locus that replaces the Rhino-specific residue with the aspartic acid residue present in all other HP1 family members. This novel allele largely phenocopies the defects in hatch rate, chromatin organization, and piRNA production associated with kipferl mutants, and does not support Kipferl localization to clusters. The data are of high quality, the presentation is clear and concise, and the conclusions are generally well-supported.

Weaknesses: 

The reviewers identified potential ways to further strengthen the manuscript.

(1) The one significant omission is RNAseq on the rhino point mutant, which would allow direct comparison to cluster, transposon, and repeat expression in kipferl mutants. 

In this eLife Advances submission, we aim to elucidate the molecular interaction between Rhino and the zinc finger protein Kipferl and how it evolved. Using various assays, of which piRNA sequencing is the most relevant and comprehensive, we show that the rhino[G31D] mutation phenocopies a rhino loss-of-function situation for Kipferl and a kipferl loss-of-function situation for Rhino. Further confirmation of this statement by additional RNA-seq experiments to probe the extent of selective TE de-repression would indeed be a possibility. We decided to test for TE de-repression phenotypes using sensitive RNA-FISH experiments of a handful of TEs that are deregulated in kipferl loss of function flies (Baumgartner at al. 2022). This showed that the same TEs are also deregulated in rhino[G31D] flies, further confirming the similarity of the two genotypes. We have added these data to the text and to Figure 5-figure supplement 2, which shows representative RNA FISH images.

(2) The manuscript would benefit from adding more evolutionary comparisons. The following or similar analyses would help put the finding into a broader evolutionary perspective:

i) Is Kipferl's surface interacting with Rhino also conserved in Kipferl orthologs? In other words, are the Rhino-interacting amino acids of Kipferl under any pressure to be conserved?

We performed an analysis of the Kipferl interface that interacts with the Rhino chromodomain in those species where Kipferl could be unambiguously identified. This showed that the residues involved in the Rhino interaction are generally conserved. We have added this analysis to Figure 1-figure supplement 4.

ii) The remarkable conservation of Rhino's G31 is at odds with the arms race that is proposed to be happening between the fly's piRNA pathway proteins and transposons. Does this mean that Rhino's chromodomain is "untouchable" for such positive selection? 

We agree that the conservation of the G31 residue argues against this binding interface being under positive selection in Rhino. Without understanding the pressures acting on Rhino that underlie the previously published positive selection, we find it difficult to draw firm conclusions. Mutating G31 in fly species that lack Kipferl would be an interesting experiment.

Recommendations for the authors:

(1) RNAseq is important to the full characterization of the phenotype and should be included. It's now clear that the major piRNA clusters are not required for fertility, so I would also include an analysis of piRNA production and Rhino binding to regions flanking isolated insertions. 

See our response to raised weakness #1 above. Briefly, we have now added an analysis of TE de-repression based on RNA-FISH experiments (Figure 5-figure supplement 2). Regarding the proposed analysis of piRNA production and Rhino binding to regions flanking isolated TE insertions: this is an important issue that we carefully analysed in our previous work characterising the kipferl mutant (Baumgartner et al. 2022). In the present work, we focused on generating a rhino mutant that uncouples Rhino from Kipferl.

(2) The authors do not provide direct biochemical evidence that the chromodomain substitution blocks Rhino binding to Kipferl. However, Rhino protein is very low abundance, making analysis of the endogenous protein very difficult.

Based on our previous work (Baumgartner et al 2022), the Rhino chromodomain interacts directly with the fourth zinc finger of Kipferl. Mutation of a single residue in the predicted interface (Rhino[G31D]) phenocopies a kipferl mutant, strongly suggesting that this mutation disrupts the Rhino-Kipferl interaction. Definitive evidence will have to await the reconstitution of this interaction using recombinant proteins. Our attempts to purify recombinant Kipferl (expressed in bacteria or in insect cells) or the protein fragments relevant to the interaction were unsuccessful so far. While we obtained soluble fractions of the first ZnF array, there was always a high level of co-purifying nucleic acids that we were not able to remove.

(3) Even if the Rhino G31D mutant retains its ability to interact with H3K9me3 it does not localize correctly on the chromatin preventing certain regions such as locus 80F from being converted into piRNA source loci. However other regions such as satellite regions attract the Rhino mutant protein converting them into super piRNA source loci, phenocopying the effects observed in a Kipferl mutant. Why Rhino when not bound to Kipferl concentrates in satellite regions is a question that remains unanswered.

This is a very interesting question indeed. We have not been able to elucidate the molecular basis of how Rhino is recruited to satellite repeats in Kipferl mutants. For example, we performed a proximity biotinylation experiment with GFP-Rhino in Kipferl mutant ovaries, but this experiment did not reveal any protein that would explain the observed accumulation of Rhino at the complex satellite repeats.

(4) In the phylogenetic analysis the authors identified two residues as Rhino-specific and conserved sequence alterations, the D31G mutation and the G62 insertion. However, the authors limit their study to D31G mutation, and nothing is performed on the G32 insertion. It would have been interesting to know the impact of this insertion on Rhino's biology. 

The role, if any, of the Rhino-specific G62 insertion and its effect on Rhino localisation or function is an interesting topic for further study. We have not investigated the G62 residue experimentally. In the current manuscript, we limited our efforts to the analysis of the G31D mutation, as the goal was to identify the mode of interaction with Kipferl, and the G62 residue is not predicted to contact Kipferl according to AlphaFold.

(5) The authors report that the G31D mutation of Rhino phenocopies the Kipferl mutant. Rhino is wrongly localized in the nucleus, and Rhino G31D recruitment in certain Kipferl-enriched regions is affected, as at the 80F locus, which correlates with a strong drop in piRNA production from this locus. To go a step further in demonstrating that G31D phenocopies the Kipferl mutant, it would have been informative to analyse how much TE piRNAs are affected and whether TEs are deregulated.

See our response to similar comments above. We have added RNA-FISH experiments to illustrate that the TE de-repression phenotypes are comparable between rhino[G31D] and kipferl loss of function ovaries (Figure 5-figure supplement 2). Analyses of TE-mapping piRNAs also show well correlated phenotypes (Figure 5-figure supplement 1).

(6) Figure 3A: To homogenize with the immunostaining presented in Figure 3B, can the authors add on the bar graph depicting female fertility the results obtained with kipferl-/- and rhino-/- genotype? 

rhino mutants are completely (100%) sterile and the fertility of kipferl mutants was previously measured to range between 15% and 40% (Baumgartner et al. 2022).

(7) Figure 4A: It would have been interesting to show Venn diagrams showing the overlap of genomic regions enriched for Kipferl versus regions enriched for Rhi in a WT and in a Rhi G31D mutant. 

We consider the analysis presented in Figure 4 to be more meaningful, as a Venn diagram would require binary cut-offs.

(8) Figure 1B: In the phylogenic analysis for Rhino/HP1d two D. simulans lines are presented. Can the authors clarify this point?

There are two Rhino paralogs in D. simulans: one paralog (NCBI: AAY34025.1) is more similar to D. melanogaster Rhino, contains one intron and is located at chromosome chr2R (assembly Apr. 2005, WUGSC mosaic 1.0/droSim1: 12256895-12258668). The second paralog (XP_002106478.1) is located on chromosome X (6734493-6735248) and does not contain an intron. We have added a clarifying statement to the corresponding figure legend.

(9) To determine whether Rhino G31D point mutation affects the overall function of Rhino, the authors analysed Kipferl-independent piRNA source loci by looking at Responder and 1,688 family satellites. I'm not sure that these loci can be classified as Kipferl-independent piRNA source loci since a strong increase of piRNA production from these loci in Kipferl mutant is observed. In my point of view, the 42AB and 38C are real Kipferl-independent piRNA source loci as piRNA production from these loci is not affected by Kipferl KD. 

Indeed, the Rsp and 1,688 family satellites are not completely independent of Kipferl, as their expression and Rhino occupancy differ between wild-type and kipferl loss-of-function phenotypes (including rhino[G31D]). However, we believe that this increase is due to a strong dependence on different sequestration mechanisms and is not mediated by a direct function of Kipferl at these sites. Similarly, we observe slight differences in piRNA production for the peripheral parts of cluster 42AB, as well as differences in Rhino occupancy despite an unaltered piRNA profile at cluster 38C (Baumgartner et al. 2022). Thus, different flavours of Kipferl-independence exist, with the only truly Kipferl-independent piRNA sources likely to be the piRNA clusters in the testis. A clear classification is further complicated by previously observed compensatory effects in the piRNA pathway, leading us to adopt the current definition of "requiring Kipferl for Rhino recruitment" to distinguish Kipferl-dependent from Kipferl-independent sites.

(10) The authors report that the G31D mutation of Rhino phenocopies the Kipferl mutant. Rhino is wrongly localized in the nucleus, and Rhino G31D recruitment in certain Kipferl-enriched regions is affected, as at 80F locus, which correlates with a strong drop in piRNA production from this locus. To go a step further in demonstrating that G31D phenocopies the Kipferl mutant, it would have been interesting to look at how much TE piRNAs are affected and whether TEs (and which class of TE) are deregulated by RNAseq and/or in situ hybridization. 

See our response to similar comments above. Our new RNA-FISH experiments and TE-mapping piRNA analysis extend the comparison of phenotypes between kipferl mutants and rhino[G31D] mutants and are consistent with our previous conclusions (Figure 5-figure supplements 1 and 2).

eLife assessment

Schafer et al. investigate the extremely interesting and important claim that the human hippocampus represents the interactions with multiple social interaction partners on two relatively abstract social dimensions – and that this ability correlates with the social network size of the participant. This research potentially demonstrates the intricate role of the hippocampus in navigating our social world. While some results are tantalizing, the empirical evidence for the main claims is currently incomplete and requires clarifications and substantial revisions.

Reviewer #1 (Public Review):

Summary:

In this study, Jellinger et al. performed engram-specific sequencing and identified genes that were selectively regulated in positive/negative engram populations. In addition, they performed chronic activation of the negative engram population over 3 months and observed several effects on fear/anxiety behavior and cellular events such as upregulation of glial cells and decreased GABA levels.

Strengths:

They provide useful engram-specific GSEA data and the main concept of the study, linking negative valence/memory encoding to cellular level outcomes including upregulation of glial cells, is interesting and valuable.

Comments on the revised manuscript:

The revised manuscript still does not adequately address the primary technical concern regarding long-term DREADD manipulation. The authors reference their previous work (Suthard et al., 2023) as evidence; however, this earlier paper only presents fluorescence intensity in a non-quantitative manner with merely three samples (Supplementary Figure 7). This limited evidence does not sufficiently support the claim of potent long-term activation. The discussion in the revision stating "...even if our manipulation is only working for 1 month, rather than 3 months..." is unconvincing, particularly given that the title and abstract still claims "chronic activation of...". To substantiate the technical validity of the study, at least cFos staining at various time points is necessary, which is less burdensome compared to more direct demonstrations such as slice physiology. Thus, although I believe it could be an interesting study for some audiences, I cannot support the strength of the evidence presented in the study.

Furthermore, in response to all reviewers' concerns regarding the quantification of GABA, the authors have removed the data from the study rather than providing properly acquired images or quantified data. This action diminishes the significance of the study.

eLife assessment

This useful study reports the behavioural and physiological effects of the longitudinal activation of neurons associated with negative experiences. The main claims of the paper are supported by solid experimental evidence, although the specificity of the long-term manipulation could have benefitted from additional validation. This study will be of interest to neuroscientists working on memory.

Reviewer #2 (Public Review):

Summary:

Jellinger, Suthard, et al. investigated the transcriptome of positive and negative valence engram cells in the ventral hippocampus, revealing anti- and pro-inflammatory signatures of these respective valences. The authors further reactivated the negative valence engram ensembles to assay the effects of chronic negative memory reactivation in young and old mice. This chronic re-activation resulted in differences in aspects of working memory, fear memory, and caused morphological changes in glia. Such reactivation-associated changes are putatively linked to GABA changes and behavioral rumination.

Strengths:

Much the content of of this manuscript is of benefit to the community, such as the discovery of differential engram transcriptomes dependent on memory valence. The chronic activation of neurons, and the resultant effects on glial cells and behavior, also provide the community with important data. Laudable points of this manuscript include the comprehensiveness of behavioral experiments, as well as the cross-disciplinary approach.

Weaknesses:

Weaknesses noted in the previous version of the manuscript have been accounted for.

Reviewer #3 (Public Review):

Summary:

The authors note that negative ruminations can lead to pathological brain states and mood/anxiety dysregulation. They test this idea by using mouse engram-tagging technology to label dentate gyrus ensembles activated during a negative experience (fear conditioning). They show that chronic chemogenetic activation of these ensembles leads to behavioral (increased anxiety, increased fear generalization, reduced fear extinction) and neural (increases in neuroinflammation, microglia and astrocytes).

Strengths:

The question the authors ask here is an intriguing one, and the engram activation approach is a powerful way to address the question. Examination of a wide range of neural and behavioral dependent measures is also a strength.

Weaknesses:

The major weakness is that the authors have found a range of changes that are correlates of chronic negative engram reactivation. However, they do not manipulate these outcomes to test whether microglia, astrocytes, neuroinflammation are causally linked to the dysregulated behaviors.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1 (Public Review):

Summary:

In this study, Jellinger et al. performed engram-specific sequencing and identified genes that were selectively regulated in positive/negative engram populations. In addition, they performed chronic activation of the negative engram population over 3 months and observed several effects on fear/anxiety behavior and cellular events such as upregulation of glial cells and decreased GABA levels.

Strengths:

They provide useful engram-specific GSEA data and the main concept of the study, linking negative valence/memory encoding to cellular level outcomes including upregulation of glial cells, is interesting and valuable.

Weaknesses:

A number of experimental shortcomings make the conclusion of the study largely unsupported. In addition, the observed differences in behavioral experiments are rather small, inconsistent, and the interpretation of the differences is not compelling.

Major points for improvement:

(1) Lack of essential control experiments

With the current set of experiments, it is not certain that the DREADD system they used was potent and stable throughout the 3 months of manipulations. Basic confirmatory experiments (e.g., slice physiology at 1m vs. 3m) to show that the DREADD effects on these vHP are stable would be an essential bottom line to make these manipulation experiments convincing.

In previous work from our lab performing long-term activation of Gq DREADD receptors in the vHPC, we quantify the presence of Gq receptor expression over 3-, 6- and 9-month timepoints and show that there is no decrease in receptor expression, as measured via fluorescence intensity (Suthard et al., 2023). In this study, we also address that even if our manipulation is only working for 1 month, rather than 3 months, we are observing the long-term effects of this shorter-term stimulation. This is still relevant, and only changes how we interpret these findings, as shorter-term stimulation or disruption of neuronal activity can still have detrimental effects on behavior.

Furthermore, although the authors use the mCherry vector as a control, they did not have a vehicle/saline control for the hM3Dq AAV. Thus, the long-term effects such as the increase in glial cells could simply be due to the toxicity of DREADD expression, rather than an induced activity of these cells.

For chemogenetic studies, our experimental rationale utilized a standard approach in the field, which includes one of two control options: 1) active receptor vs. control vector + ligand or 2) active receptor + ligand or saline control. We chose the first option, as this more properly controls for the potential off-target effects of the ligand itself, as shown in other previous work (Xia et al., 2017). This is particularly important for studies using CNO, as many off-target effects have been noted as a limitation (Manvich et al., 2018). We chose to use DCZ as it is closely related to CNO and newer ligands, but comes with added benefits of high specificity, low off-target effects, high potency and brain penetrance (Nagai et al., 2020), but any potential off-target effects of DCZ are yet to be completely investigated as this ligand is very new.

Evidence of DREADD toxicity has been shown at high titer levels of AAV2/7- CamKIIα-hM4D(Gi)-mCherry in the hippocampus at 5 weeks, as the reviewer pointed out in their above comment (Goossens et al., 2021). Our viral strategy is targeted to a much smaller number of cells using AAV9-DIO-Flex-hM3Dq-mCherry at a lower titer, unlike expression within a much larger population of CaMKII+ excitatory neurons in this study. Additionally, visual comparison of their viral load and expression with ours shows much more intense expression that spans a larger area of the hippocampus (Goossens et al, 2021; Figure 1D), whereas ours is isolated to a smaller region of vHPC (see Figure 1B).

Further, we attempted to quantify a decrease in neuronal health (Yousef et al., 2017) resulting from DREADD expression via NeuN counts within multiple hippocampal subregions for the 6- and 14-month groups across active Gq receptor and mCherry conditions and did not observe significant decreases in NeuN as a result (Supplemental Figure 1). However, immunohistochemistry of an individual marker may not be sufficient to capture the entire health profile of an individual neuron and future work should consider other markers of cell death or inflammation, which we have added to the Limitations & Future Work section of our Discussion.

(2) Figure 1 and the rest of the study are disconnected

The authors used the cFos-tTA system to label positive/negative engram populations, while the TRAP2 system was used for the chronic activation experiments. Although both genetic tools are based on the same IEG Fos, the sensitivity of the tools needs to be validated. In particular, the sensitivity of the TRAP2 system can be arbitrarily altered by the amount of tamoxifen (or 4OHT) and the administration protocols. The authors should at least compare and show the percentage of labeled cells in both methods and discuss that the two experiments target (at least slightly) different populations. In addition, the use of TRAP2 for vHP is relatively new; the authors should confirm that this method actually captures negative engram populations by checking for reactivation of these cells during recall by overlap analysis of Fos staining or by artificial activation.

We thank the reviewer for their comments and opportunity to discuss the marked differences between TRAP2 and DOX systems. In particular, we agree that while both systems rely on the the Fos promoter to drive an effector of interest, their efficacy and temporal resolution vary substantially depending on genetic cell-type, brain region, temporal parameters of Dox or 4-OHT delivery, subject-by-subject metabolic variability, and threshold to Fos induction given the promoter sequences inherent to each system. For example, recent studies have reported the following:

- The TRAP2 line labels a subset of endogenously activeCA1 pyramidal cells (e.g. 5-18%) while the DOX system labels 20-40% of CA1 pyramidal cells (DeNardo et al, 2019; Monasterio et al, BioRxiv 2024 ).

- The temporal windows for each range from hours in TRAP2 to 24-48 hours for DOX (DeNardo et al, 2019; Denny et al, 2014; Liu & Ramirez et al, 2012).

- The efficacy of “tagging” a population of cells with TRAP2 vs with DOX will constrain the number of possible cells that may overlap with cFos upon re-exposure to a given experience (e.g. see the observed overlaps in vCA1 - BLA circuits (Kim & Cho, 2020), compared to vCA1 in general (Ortega-de San Luis et al, 2023) and valence-specific vCA1 populations (Shpokayte et al, 2022).

- Tagging vCA1 cells with both the TRAP2 and DOX systems are nonetheless sufficient to drive corresponding behaviors (e.g. vCA1 terminal stimulation drives behavioral changes with the DOX and TRAP2 system (Shpokayte et al, 2022) and vCA1 stimulation of an updated fear-linked ensemble drives light-induced freezing in a neutral context utilizing the TRAP2 and DOX systems (Ortega-de San Luis et al, 2023)).

Finally, and promisingly, as more studies continue to link the in vivo physiological dynamics of these cell populations tagged using each system (e.g. compare Pettit et al, 2022 with Tanaka et al, 2018) and correlating their activity to behavioral phenotypes, our field is in the prime position to uncover deeper principles governing hippocampus-mediated engrams in the brain. Together, we believe a more comprehensive understanding of these systems is fully warranted, especially in the service of further cataloging cellular similarities and differences within such tagged populations.

(3)  Interpretation of the behavior data

In Figures 3a and b, the authors show that the experimental group showed higher anxiety based on time spent in the center/open area. However, there were no differences in distance traveled and center entries, which are often reduced in highly anxious mice. Thus, it is not clear what the exact effect of the manipulation is. The authors may want to visualize the trajectories of the mice's locomotion instead of just showing bar graphs.

Our findings show that our experimental group displays higher levels of anxiety-like behaviors as measured via time spent in center/open area, while there are no differences in distance traveled or center entries. For distance traveled, our interpretation is in line with complementary research (Jimenez et al, 2018; Kheirbek et al, 2013) that shows no changes in distance traveled/distance traveled in the center coupled with changes in anxiety levels as a result of manipulation within anxiety-related circuits. More broadly, any locomotion-related deficit could cause a change in distance traveled that is unrelated to anxiety-like behaviors alone. For example, a reduction in distance traveled could be coupled with a decrease in time spent in the center, but could also result only from motor or exploratory deficits. We hope that this explanation clarifies our interpretation of the open field and elevated plus maze findings in light of other literature.

In addition, the data shown in Figure 4b is somewhat surprising - the 14MO control showed more freezing than the 6MO control, which can be interpreted as "better memory in old". As this is highly counterintuitive, the authors may want to discuss this point. The authors stated that "Mice typically display increased freezing behavior as they age, so these effects during remote recall are expected" without any reference. This is nonsense, as just above in Figure 4a, older mice actually show less freezing than young mice. Overall, the behavioral effects are rather small and random. I would suggest that these data be interpreted more carefully.

In Figure 4B, we present our findings from remote recall and observe increased freezing levels in control mice with age, as mentioned by the reviewer, indicating increased memory. This is in line with previous work from Shoji & Miyakawa, 2019 which has been added as a reference for the quotation described above; we thank the reviewer for pointing this error out. As the reviewer has pointed out, above in Figure 4A, we measured freezing levels across all groups during contextual fear conditioning before the start of chronic stimulation, as this was the session we ‘tagged’ a negative memory in. Although it appears that there may be slightly lower levels of freezing in older (14-month old) mice, our findings do not determine statistical significance for difference between age group, only effects of time and subject which are expected as freezing increases within the session and animals display high levels of variability in freezing levels across many experiments (Figure 4A i-iii). We also find in previous work that control mice receiving 3-, 6- and 9-months of chronic DCZ stimulation in the vHPC with empty vector (mCherry) receptor show an increase in freezing with age (Suthard et al, 2023; Figure 2A ii).

(4) Lack of citation and discussion of relevant study

Khalaf et al. 2018 from Gräff lab showed that experimental activation of recall-induced populations leads to fear attenuation. Despite the differences in experimental details, the conceptual discrepancy should be discussed.

As mentioned by the reviewer, Khalaf et al. 2018 showed that experimental activation of recall-induced populations in the dentate gyrus leads to fear attenuation. Specifically, they pose that this fear attenuation occurs in these ensembles through updating or unlearning of the original memory trace via the engagement, rather than suppression, of an original traumatic experience. Despite the differences in experimental details with our current study and this work, we agree that the conceptual discrepancy should be discussed. First, one major difference is that we are reactivating an ensemble that was tagged during fear memory encoding, while Khalaf et al. are activating a remote recall-induced ensemble that was tagged one month after encoding. Although there is high overlap between the encoding and recall ensembles when mice are exposed to the conditioning context, these ensembles are not identical and may result in different behavioral phenotypes when chronically reactivated. Further, Khalaf et al rely on reactivation of the recall-induced ensemble during extinction to facilitate rapid fear attenuation. This differs from our current work, as their reactivation is occurring during the extinction process in the previously conditioned context, while we are reactivating chronically in the animal’s home cage over the course of a longer time period. It may be necessary that the memory is first reactivated, and thus, more liable to re-contextualization, in the original context compared to an unrelated homecage environment where there are presumably no related cues present. Importantly, this previous work tests the attenuation of fear shortly after an extinction process, while we are not traditionally extinguishing the context with aid of the memory reactivation. Finally, we are testing remote recall (3 months post-conditioning), while they are testing at a shorter time interval (28 days). In line with these ideas, future work may seek to tease out the mechanistic differences between recent and remote memory extinction both in terms of natural memory recall and chronically manipulated memory-bearing cells.

Reviewer #2 (Public Review):

Summary:

Jellinger, Suthard, et al. investigated the transcriptome of positive and negative valence engram cells in the ventral hippocampus, revealing anti- and pro-inflammatory signatures of these respective valences. The authors further reactivated the negative valence engram ensembles to assay the effects of chronic negative memory reactivation in young and old mice. This chronic re-activation resulted in differences in aspects of working memory, and fear memory, and caused morphological changes in glia. Such reactivation-associated changes are putatively linked to GABA changes and behavioral rumination.

Strengths:

Much of the content of this manuscript is of benefit to the community, such as the discovery of differential engram transcriptomes dependent on memory valence. The chronic activation of neurons, and the resultant effects on glial cells and behavior, also provide the community with important data. Laudable points of this manuscript include the comprehensiveness of behavioral experiments, as well as the cross-disciplinary approach.

Weaknesses:

There are several key claims made that are unsubstantiated by the data, particularly regarding the anthropomorphic framing of "rumination" on a mouse model and the role of GABA. The conclusions and inferences in these areas need to be carefully considered.

(1) There are many issues regarding the arguments for the behavioural data's human translation as "rumination." There is no definition of rumination provided in the manuscript, nor how rumination is similar/different to intrusive thoughts (which are psychologically distinct but used relatively interchangeably in the manuscript), nor how rumination could be modelled in the rodent. The authors mention that they are attempting to model rumination behaviours by chronically reactivating the negative engram ("To understand if our experimental model of negative rumination..."), but this occurs almost at the very end of the results section, and no concrete evidence from the literature is provided to attempt to link the behavioural results (decreased working memory, increased fear extinction times) to rumination-like behaviours. The arguments in the final paragraph of the Discussion section about human rumination appear to be unrelated to the data presented in the manuscript and contain some uncited statements. Finally, the rumination claims seem to be based largely upon a single data figure that needs to be further developed (Figure 6, see also point 2 below).

(2) The staining and analysis in Figure 6 are challenging to interpret, and require more evidence to substantiate the conclusions of these results. The histological images are zoomed out, and at this resolution, it appears that only the pyramidal cell layer is being stained. A GABA stain should also label the many sparsely spaced inhibitory interneurons existing across all hippocampal layers, yet this is not apparent here. Moreover, both example images in the treatment group appear to have lower overall fluorescence intensity in both DAPI and GABA. The analysis is also unclear: the authors mention "ROIs" used to measure normalized fluorescence intensity but do not specify what the ROI encapsulates. Presumably, the authors have segmented each DAPI-positive cell body and assessed fluorescence however, this is not explicated nor demonstrated, making the results difficult to interpret.

Based on the collective discussion from all reviewers on the completeness of our GABA quantification and its implications, we have decided to remove this figure and perform more substantive analysis of this E/I imbalance in future work.

(3) A smaller point, but more specific detail is needed for how genes were selected for GSEA analysis. As GSEA relies on genes to be specified a priori, to avoid a circular analysis, these genes need to be selected in a blind/unbiased manner to avoid biasing downstream results and conclusions. It's likely the authors have done this, but explicitly noting how genes were selected is an important context for this analysis.

As mentioned in our Methods section, gene sets were selected based on pre-existing biology and understanding of genes canonically involved in “neurodegeneration” such as those related to apoptotic pathways and neuroinflammation or “neuroprotection” such as brain-derived neurotrophic factor, to name a few. A limitation of this method is that we must avoid making strong claims about the actual function of these up- or down-regulated genes without performing proper knock-in or knock-out studies, but we hope that this provides an unbiased inventory for future experiments to perform causal manipulations.

Reviewer #3 (Public Review):

Summary:

The authors note that negative ruminations can lead to pathological brain states and mood/anxiety dysregulation. They test this idea by using mouse engram-tagging technology to label dentate gyrus ensembles activated during a negative experience (fear conditioning). They show that chronic chemogenetic activation of these ensembles leads to behavioral (increased anxiety, increased fear generalization, reduced fear extinction) and neural (increases in neuroinflammation, microglia, and astrocytes).

Strengths:

The question the authors ask here is an intriguing one, and the engram activation approach is a powerful way to address the question. Examination of a wide range of neural and behavioral dependent measures is also a strength.

Weaknesses:

The major weakness is that the authors have found a range of changes that are correlates of chronic negative engram reactivation. However, they do not manipulate these outcomes to test whether microglia, astrocytes, or neuroinflammation are causally linked to the dysregulated behaviors.

Recommendations For The Authors:

Reviewer #1 (Recommendations For The Authors):

- Figure 2c should include Month0, the BW before the start of the manipulation.

Regrettably, we do not have access to the Month 0 body weights at this time as this project changed hands over the course of the past year or so. This is an inherent limitation that we missed during analysis and we pose this as a limitation in the Results section after describing this finding. Therefore, it is possible that over the first month of stimulation (Month 0-1), there may have been a drop in body weight that rebounded by the first measurement at Month 1 that continued to increase normally through Months 2-3, as shown in our Figure 1. Thank you for this note.

- Figure 6a looks confusing - the background signal in the green channel is very different between control and experimental groups. Were representative images taken with different microscope settings?

The representative images were taken with the same microscope power settings, but were adjusted in brightness/contrast within FIJI for clarity in the Figure – we apologize that this was misleading in any way and thank the reviewer for their feedback. Further, based on the collective discussion from all reviewers on the completeness of our GABA quantification and its implications, we have decided to remove this figure and perform more substantive analysis of this E/I imbalance in future work.

- Typo mChe;try

This typo was fixed

- "During this contextual... mice in the 6- and 14- month groups..." Isn't it 3- and 11- month respectively at the time of fear conditioning? Throughout the manuscript, this point was written very confusingly.

Yes, we thank the reviewer for pointing this out. It has been corrected to 3- and 11-month old mice at the timing of fear conditioning and clarified throughout the manuscript where applicable.

- "GABAergic eYFP fluorescence" Where does the eYFP come from? The methods state that GABA quantification is based on IHC staining.

Based on the collective discussion from all reviewers on the completeness of our GABA quantification and its implications, we have decided to remove this figure and perform more substantive analysis of this

E/I imbalance in future work. We discuss this E/I balance not being directly assessed in the Limitations & Future Directions section of our Discussion, noting the importance of detailed quantification of both excitatory and inhibitory markers within the hippocampus.

Reviewer #2 (Recommendations For The Authors):

(1) There is a full methods section ("Analysis of RNA-seq data") that mostly describes RNA-seq analysis that seemingly does not appear in the paper. This section should be reviewed.

We have included this portion of the methods that explain the previous workflow from Shpokayte et al., 2022 where this dataset was generated and this has been noted in the “Analysis of RNA-seq data” section of the methods.

(2) Figure 6: GABA staining should be more critically analyzed, as discussed above, and validated with another GABA antibody for rigor. From the representative images provided in Figure 6, it looks possibly as though the hM3Dq images were simply not fully in the focal plane when being imaged or were over-washed, as DAPI staining also appears to be lower in these images.

Based on the collective discussion from all reviewers on the completeness of our GABA quantification and its implications, we have decided to remove this figure and perform more substantive analysis of this E/I imbalance in future work. Specifically, it will be necessary to rigorously investigate both excitatory and inhibitory markers within this region to ensure these claims are substantiated. Thank you for this suggestion.

(3) The first claim that human GABAergic interneurons cause rumination is uncited. (Page 19, first sentence beginning with: "Evidence from human studies suggests...").

Based on the collective discussion from all reviewers on the completeness of our GABA quantification and its implications, we have decided to remove this figure and perform more substantive analysis of this E/I imbalance in future work. Apologies for the lack of citation in-text, the proper citation for this finding is Schmitz et al, 2017.

(4) Gene names throughout the manuscript and figure are written in the wrong format for mice (eg: Page 13, second line: SPP1, TTR, and C1QB1 instead of Spp1, Ttr, C1qb1).

This was corrected throughout the manuscript.

(5) Tense on Page 15 third sentence of the second paragraph: "...spatial working memory was assessed...".

This was corrected throughout the manuscript.

(6) Supplemental Figure 1 would benefit from normalization of the NeuN+ cell counts. The inclusion of an excitatory and inhibitory neuron marker in this figure might benefit the argument that there is a change in the excitation/inhibition of the hippocampus - as the numbers of excitatory neurons outweigh the numbers of inhibitory neurons that would be assayed here.

In an effort to normalize the NeuN+ cell counts, for each of our ROIs (6-8 single tiles for each brain region (DG, vCA1, vSub) x 3-5 coronal slices = ~18 single tiles per mouse x 3-4 mice) we captured a 300 x 300 micrometer, single-tile z-stack at 20x magnification. These ROIs were matched for dimensions and brain regions across all groups for each hippocampal subregion quantified. We initially proposed to normalize these NeuN counts over DAPI, but because DAPI includes all nuclei (microglia, oligodendrocytes, astrocytes and neurons), we weren’t sure this was the most optimal tool. We do agree that further quantification of excitatory and inhibitory cell markers would be vital to more concrete interpretation of our findings and we have added this to our Limitations & Future Work section of the Discussion.

Reviewer #3 (Recommendations For The Authors):

(1) The DOX tagging window lacks temporal precision. I suggest the authors note this as a limitation.

We thank the reviewer for noting this, and we have added this limitation to the Methods section with the context of the 24-48 hour DOX window being longer than other methods like TRAP.

(2) Is there a homeostatic response to chronic engram stimulation? That is, is DCZ as effective in increasing neuronal excitability on day 90 as it is on day 1. This could be addressed with electrophysiology, or with IEG induction. Alternatively, the authors could refer to previous literature-- for example, Xia et al (2017) eLife-- that examined whether there was any blunting of the effects of DREADD ligands after sustained delivery via drinking water. There, of course, may be other papers as well.

As noted by the reviewer, it is important to determine if DCZ maintains its effects on neuronal excitability throughout the 3 month administration period. To address this, previous work has shown that CNO administration in drinking water over one month consistently inhibited hM4Di+ neurons without altering baseline neuronal excitability as measured by firing rate and potassium currents (Xia et al, 2017). Although this is only for one month, it is administered via the same oral route as our DCZ protocol and suggests that at least for that amount of time we are likely producing consistent effects. In our reply above to Reviewer #1’s comment, we also note that even if DCZ is only having an effect for one month, rather than 3 months, we are still observing enduring changes that resulted from this short-term disturbance.

(3) Please double check there is no group effect on weight in 6-month-old mice in Figure 2C.

Two-way RM ANOVA showed no main effect of Group within the 6-month-old control and hM3Dq groups.

Group: F(1,17) = 1.361, p=0.2594.

(4) The shock intensity is much higher than is typical for fear conditioning studies in mice. Why was this the case?

Yes, we do agree that this shock intensity is on the higher side of typical paradigms in mice, however, our lab has utilized 0.75mA to 1.5mA intensity foot shocks for contextual fear conditioning in the past (Suthard & Senne et al, 2023; 2024; Dorst & Senne et al, 2023; Grella et al., 2022; Finkelstein et al., 2022) and we maintained this protocol for internal consistency. However, it would be interesting to systematically investigate how differing intensities of foot shock, subsequent tagging of this ensemble and reactivation would uniquely impact behavioral state acutely and chronically in mice.

(5) Remote freezing is very low. The authors should comment on this-- perhaps repeated testing has led to some extinction?

A reviewer above suggested a similar phenomenon may be occuring, specifically fear attenuation as a result of chronic stimulation. They referenced previous work from Khalaf et al. 2018, where they reactivated a recall-induced ensemble, while we reactivated an ensemble tagged during encoding. We expand upon this work in light of our findings within the Limitations & Future Work section of our Discussion. However, we do appreciate the lower levels of freezing observed in remote recall and sought out other literature to understand the typical range of remote freezing levels. One thing that we note is that our remote recall is occurring 3 months after conditioning, which is much longer than typical 14-28 day protocols. However, we find that freezing levels at remote timepoints from 21-45 days results in contextual freezing levels of between 20-50% approximately (Kol et al., 2020), as well as 40-75% approximately in a variety of 28 day remote recall experiments (Lee et al., 2023). This information, together with our current experimental protocol demonstrates a wide range of remote freezing levels that may depend heavily on the foot shock intensity, duration of days after conditioning, and animal variability.

(6) "mice display increased freezing with age": please add a reference.

Apologies, we missed the citation for that claim and it has been added in-text and in the references list (Shoji & Miyakawa, 2019).

(7) Related to the low freezing levels for remote memory, why is generalization minimal? Many studies have shown that there is a time-dependent emergence of generalized fear, yet here this is not seen. Is it linked to extinction (as above)? Or genetic background?

Previous work has shown that rats receiving multiple foot shocks during conditioning displayed a time-dependent generalization of context memory, while those receiving less shocks did not (Poulos et al., 2016), as the reviewer noted in their comment. In our current study, we observe low levels of generalization in all of our groups compared to freezing levels displayed in the conditioned context at the remote timepoint, in opposition to this time-dependent enhancement of generalization. It is possible that the genetic background of our C57BL/6J mice compared to the Long-Evans rat strain in this previous work accounts for some of this difference. In addition, it is possible that the longer duration of time (3 months) compared to their remote timepoint (28 days) resulted in time-dependent decrease in generalization that decreases with greater durations of time from original conditioning. As noted above, it is indeed plausible that the reactivation of a contextual fear ensemble over time is attenuating freezing levels for both the original and similar contexts (Khalaf et al, 2018). We discuss the differences in our study and this 2018 work more comprehensively above.

(8) Morphological phenotypes of astrocytes/microglia. Would be great to do some transcriptomic profiling of microglia/astrocytes to couple with the morphological characterization (but appreciate this is beyond the scope of current work).

We thank the reviewer this suggestion, we agree that would be an incredibly informative future experiment and have added this to our Limitations & Future Experiments section of the Discussion.

(9) The authors could consider including a limitations section in their discussion which discusses potential future directions for this work:

- causal experiments.

- E/I balance is not assessed directly (interestingly, in this regard, expanded engrams are linked to increased generalization [e.g., Ramsaran et al 2023]).

Thank you for this suggestion, we have added a Limitations & Future Directions section to our Discussion and have expanded upon these suggested points.

(10) For Figure 10, consider adding an experimental design/timeline.

We are making the assumption that the reviewer meant Figure 1 instead of Figure 10 here, but note that there is a description of the viral expression duration (D0-D10), followed by an off Dox period of 48 hours (D10-D12), with subsequent engram tagging of a negative (foot shock) or positive (male-to-female exposure) on D12. In our experiments (Shpokayte et al., 2022), Dox was administered for 24 hours (D12-D13), which was followed by sacrificing the animal for cell suspension and sequencing of the positive and negative engram populations. This figure also shows the viral strategy for the Tet-tag system (Figure 1A), as well as representative viral expression in vHPC (Figure 1B). We are happy to add additional experimental design/timeline information to this figure that would be helpful to the reviewer.

https://www.derstandard.at/story/3000000226570/gerichtshof-fuer-menschenrechte-raeumt-klimaklage-gegen-oesterreich-prioritaet-ein

Let's break down how to create a hyperlinked API in Django REST framework, step by step, with an example.

What is a Hyperlinked API?

A hyperlinked API means that instead of using primary keys to reference related objects, we use URLs (hyperlinks). This makes the API more intuitive and easier to navigate.

Steps to Create a Hyperlinked API

1. Update Serializers:
2. Use HyperlinkedModelSerializer instead of ModelSerializer.

3. Add URL fields to represent relationships as hyperlinks.

4. Update URL Patterns:

5. Name the URL patterns so that they can be referenced by the serializers.

Example

Step 1: Update Serializers

In snippets/serializers.py, update your serializers to use HyperlinkedModelSerializer:

```python from rest_framework import serializers from .models import Snippet from django.contrib.auth.models import User

class SnippetSerializer(serializers.HyperlinkedModelSerializer): owner = serializers.ReadOnlyField(source='owner.username') highlight = serializers.HyperlinkedIdentityField(view_name='snippet-highlight', format='html')

class Meta:
    model = Snippet
    fields = ['url', 'id', 'highlight', 'owner', 'title', 'code', 'linenos', 'language', 'style']

class UserSerializer(serializers.HyperlinkedModelSerializer): snippets = serializers.HyperlinkedRelatedField(many=True, view_name='snippet-detail', read_only=True)

class Meta:
    model = User
    fields = ['url', 'id', 'username', 'snippets']

```

Explanation: - SnippetSerializer: - owner: Read-only field that shows the username of the snippet owner. - highlight: Hyperlinked field pointing to the 'snippet-highlight' URL. - fields: List of fields to include in the serialized output.

• UserSerializer:
• snippets: Hyperlinked field that shows all snippets owned by the user, pointing to the 'snippet-detail' URL.
• fields: List of fields to include in the serialized output.

Step 2: Update URL Patterns

In snippets/urls.py, name your URL patterns:

```python from django.urls import path from rest_framework.urlpatterns import format_suffix_patterns from snippets import views

API endpoints

urlpatterns = format_suffix_patterns([ path('', views.api_root), path('snippets/', views.SnippetList.as_view(), name='snippet-list'), path('snippets/<int:pk>/', views.SnippetDetail.as_view(), name='snippet-detail'), path('snippets/<int:pk>/highlight/', views.SnippetHighlight.as_view(), name='snippet-highlight'), path('users/', views.UserList.as_view(), name='user-list'), path('users/<int:pk>/', views.UserDetail.as_view(), name='user-detail') ]) ```

Explanation: - format_suffix_patterns: Allows adding format suffixes like .json or .html to URLs. - path: Defines URL patterns and associates them with views. - name: Names the URL patterns so that they can be referenced by the serializers.

How It Works

1. Request: When a user requests a snippet or user detail, the serializer returns URLs for related objects instead of primary keys.
2. Navigation: The user can follow these URLs to navigate between related objects.

Example in Action

1. User Request: GET /snippets/
2. Response: json [ { "url": "http://example.com/snippets/1/", "id": 1, "highlight": "http://example.com/snippets/1/highlight/", "owner": "user1", "title": "Example Snippet", "code": "print('Hello, World!')", "linenos": true, "language": "python", "style": "friendly" } ]

Here, the owner field is a username, and the highlight and url fields are hyperlinks to the related endpoints.

This is how you create a hyperlinked API using Django REST framework, making it easier to navigate relationships between entities.

Let's break down how to create an endpoint for code highlighting in a simple way, with an example:

What is an Endpoint?

An endpoint is a specific URL where our web application can send requests to get or send data. In this case, we want to create an endpoint to highlight code snippets and return an HTML representation instead of JSON.

Steps to Create the Endpoint

1. Create the View:
2. We will create a view called SnippetHighlight that will handle requests to highlight a code snippet.
3. This view will use a special renderer to return HTML instead of JSON.

4. Since there's no built-in view that fits our need exactly, we will create a custom view by extending GenericAPIView.

5. Update URLs:

6. We will add a new URL pattern to link to our SnippetHighlight view.

Example

Step 1: Create the View

First, we create our custom view in snippets/views.py:

```python from rest_framework import generics, renderers from rest_framework.response import Response from .models import Snippet

class SnippetHighlight(generics.GenericAPIView): queryset = Snippet.objects.all() renderer_classes = [renderers.StaticHTMLRenderer]

def get(self, request, *args, **kwargs):
    snippet = self.get_object()
    return Response(snippet.highlighted)

```

Explanation: - Import Statements: We import necessary modules. - SnippetHighlight Class: This class handles the requests to highlight snippets. - queryset: Specifies which snippets are available. - renderer_classes: Tells Django to use HTML renderer instead of JSON renderer. - get() Method: This method handles GET requests. It fetches the requested snippet and returns its highlighted HTML.

Step 2: Update URLs

Next, we add the URL pattern in snippets/urls.py:

```python from django.urls import path from . import views

urlpatterns = [ path('', views.api_root), # Your API root path('snippets/<int:pk>/highlight/', views.SnippetHighlight.as_view()), # URL for highlighting ] ```

Explanation: - path('', views.api_root): This is the root of our API. - path('snippets/<int:pk>/highlight/', views.SnippetHighlight.as_view()): This URL pattern connects to our SnippetHighlight view. <int:pk> is a placeholder for the snippet's ID.

How It Works

1. Request: A user sends a GET request to /snippets/1/highlight/ to highlight snippet with ID 1.
2. View: The SnippetHighlight view handles the request. It fetches the snippet with ID 1, gets its highlighted HTML, and returns it.
3. Response: The user receives the highlighted HTML of the snippet.

Example in Action

1. User Request: GET /snippets/1/highlight/
2. Backend Processing:
3. The view fetches snippet 1 from the database.
4. It gets the highlighted HTML of snippet 1.
5. Response: The user gets the HTML representation of the highlighted code snippet.

This is how you create an endpoint to highlight code snippets and return HTML using Django REST Framework.

Another Example: Using Hyperlinks for Relationships in APIs

Current State: Using Primary Keys for Relationships

Let's consider an API for managing books and authors. Currently, it looks like this:

• Author object: json { "id": 1, "name": "Jane Austen", "books": [1, 2] }

• Book object: json { "id": 1, "title": "Pride and Prejudice", "author": 1 }

Here, "books" in the author object and "author" in the book object are represented by IDs. This setup is not very user-friendly because you can't directly access the related resources.

Desired State: Using Hyperlinks for Relationships

We want to replace these IDs with URLs that point to the actual resources, making the API more intuitive and easier to navigate.

Creating a Root Endpoint

We currently have separate endpoints for "books" and "authors," but no main page that links to both. We'll create a root endpoint that serves as a starting point, listing links to these sections.

Steps to Implement the Entry Point

1. Import necessary tools: python from rest_framework.decorators import api_view from rest_framework.response import Response from rest_framework.reverse import reverse

2. Define the root view: python @api_view(['GET']) def api_root(request, format=None): return Response({ 'authors': reverse('author-list', request=request, format=format), 'books': reverse('book-list', request=request, format=format) })

3. Explanation:

4. @api_view(['GET']): This decorator makes the function a view that responds to GET requests.
5. reverse(): This function generates complete URLs for the specified endpoints ("author-list" and "book-list").
6. Response: This function returns a response containing these URLs.

Example

When you visit the root endpoint (e.g., /api/), you get a response like this:

json { "authors": "http://example.com/api/authors/", "books": "http://example.com/api/books/" }

Now, instead of dealing with IDs, you can click on the links to view the list of authors or books.

Transforming Relationships to Hyperlinks

• Author object (with hyperlinks): json { "id": 1, "name": "Jane Austen", "books": [ "http://example.com/api/books/1/", "http://example.com/api/books/2/" ] }

• Book object (with hyperlink): json { "id": 1, "title": "Pride and Prejudice", "author": "http://example.com/api/authors/1/" }

Summary

• Before: Relationships are shown using IDs (not very user-friendly).
• After: Relationships are shown using hyperlinks (much easier to navigate and understand).
• Root Endpoint: Provides a main entry point with links to "authors" and "books". This makes it easy for users to find and explore related resources in the API.

前は「ログイン」となってました

訳したほうがよさそう

案

ユーザー登録

learning_logsのテンプレートではボタン名は体言止めだったので、ここも「登録」としてはいかがでしょうか

画像を日本語に差し替えるときにはこの記事も訳す必要があると思います。

原文

The bishops and knights are good pieces to have out in the opening phase of the game. They're both powerful enough to be useful in attacking your opponent, but not so powerful that you can't afford to lose them in an early trade.

案

ビショップとナイトは、ゲームの初期段階で持っておくのに適した駒です。どちらも十分強力で敵を攻撃するのに役立ちますが、早期の駒のやり取りで失うわけにはいかないほど強力ではありません。

この言葉は英文のどこから来てますか?

常に、の方が読みやすいと思いました

表現が回りくどいので、「無効な状態では」とかはどうでしょう

アクティベート、が一般的かなと。 「アクティベイト」だと会社名がひっかかる

processを作業と訳しているのが気になる。作業だとなんかあんま考えないでできそうなふんいき。カタカナで「プロセス」でもよいのでは

訳したほうがよさそう

代案

記事一覧

訳したほうがよさそう

代案

トピック

ここもイタリックではなく太字にしてください

eLife assessment

This fundamental work proposes a novel mechanism for memory consolidation where short-term memory provides a gating signal for memories to be consolidated into long-term storage. The work combines extensive analytical and numerical work applied to three different scenarios and provides a convincing analysis of the benefits of the proposed model, although some of the analyses are limited to the type of memory consolidation the authors consider (and don't consider), which limits the impact. The work will be of interest to neuroscientists and many other researchers interested in the mechanistic underpinnings of memory.

Reviewer #2 (Public Review):

Summary:

In the manuscript the authors suggest a computational mechanism called recall-gated consolidation, which prioritizes the storage of previously experienced synaptic updates in memory. The authors investigate the mechanism with different types of learning problems including supervised learning, reinforcement learning, and unsupervised auto-associative memory. They rigorously analyse the general mechanism and provide valuable insights into its benefits.

Strengths:

The authors establish a general theoretical framework, which they translate into three concrete learning problems. For each, they define an individual mathematical formulation. Finally, they extensively analyse the suggested mechanism in terms of memory recall, consolidation dynamics, and learnable timescales.

The presented model of recall-gated consolidation covers various aspects of synaptic plasticity, memory recall, and the influence of gating functions on memory storage and retrieval. The model's predictions align with observed spaced learning effects.

The authors conduct simulations to validate the recall-gated consolidation model's predictions, and their simulated results align with theoretical predictions. These simulations demonstrate the model's advantages over consolidating any memory and showcase its potential application to various learning tasks.

The suggestion of a novel consolidation mechanism provides a good starting point to investigate memory consolidation in diverse neural systems and may inspire artificial learning algorithms.

Weaknesses:

I appreciate that the authors devoted a specific section to the model's predictions, and point out how the model connects to experimental findings in various model organisms. However, the connection is rather weak and the model needs to make more specific predictions to be distinguishable from other theories of memory consolidation (e.g. those that the authors discuss) and verifiable by experimental data.

The model is not compared to other consolidation models in terms of performance and how much it increases the signal-to-noise ratio. It is only compared to a simple STM or a parallel LTM, which I understand to be essentially the same as the STM but with a different timescale (so not really an alternative consolidation model). It would be nice to compare the model to an actual or more sophisticated existing consolidation model to allow for a fairer comparison.

The article is lengthy and dense and it could be clearer. Some sections are highly technical and may be challenging to follow. It could benefit from more concise summaries and visual aids to help convey key points.

Reviewer #3 (Public Review):

Summary:

In their article Jack Lindsey and Ashok Litwin-Kumar describe a new model for systems memory consolidation. Their idea is that a short-term memory acts not as a teacher for a long-term memory - as is common in most complementary learning systems -, but as a selection module that determines which memories are eligible for long term storage. The criterion for the consolidation of a given memory is a sufficient strength of recall in the short term memory.

The authors provide an in-depth analysis of the suggested mechanism. They demonstrate that it allows substantially higher SNRs than previous synaptic consolidation models, provide an extensive mathematical treatment of the suggested mechanism, show that the required recall strength can be computed in a biologically plausible way for three different learning paradigms, and illustrate how the mechanism can explain spaced training effects.

Strengths:

The suggested consolidation mechanism is novel and provides a very interesting alternative to the classical view of complementary learning systems. The analysis is thorough and convincing.

Weaknesses:

The main weakness of the paper is the equation of recall strength with the synaptic changes brought about by the presentation of a stimulus. In most models of learning, synaptic changes are driven by an error signal and hence cease once the task has been learned. The suggested consolidation mechanism would stop at that point, although recall is still fine. The authors should discuss other notions of recall strength that would allow memory consolidation to continue after the initial learning phase. Aside from that, I have only a few technical comments that I'm sure the authors can address with a reasonable amount of work.

Author response:

The following is the authors’ response to the original reviews.

In light of some reviewer comments requesting more clarity on the relationship between our model and prior theoretical studies of systems consolidation, we propose a modification to the title of our manuscript: “Selective consolidation of learning and memory via recall-gated plasticity.” We believe this title better reflects the key distinguishing feature of our model, that it selectively consolidates only a subset of memories, and also highlights the model’s applicability to task learning as well as memory storage.

Major comments:

Reviewer #3’s primary concern with the paper is the following: “The main weakness of the paper is the equation of recall strength with the synaptic changes brought about by the presentation of a stimulus. In most models of learning, synaptic changes are driven by an error signal and hence cease once the task has been learned. The suggested consolidation mechanism would stop at that point, although recall is still fine. The authors should discuss other notions of recall strength that would allow memory consolidation to continue after the initial learning phase.”

We thank the reviewer for drawing attention to this issue, which primarily results from a poor that memories should be interpreted as actual synaptic weight updates,∆𝑤and thus in the context choice of notation on our part. Our decision to denote memories as gives the impression of supervised learning would go to zero when the task is learned. However, in the formalism of our model, memories are in fact better interpreted as target values of synaptic weights, and the synaptic model/plasticity rule is responsible for converting these target values into synaptic weight updates. We were unclear on this point in our initial submission, because our paper primarily considers binary synaptic weights, where target synaptic weights have a one-to-one correspondence with candidate synaptic weight updates. We have updated the paper to use w* to refer to memories, which we hope resolves this confusion, and have updated our introduction to the term “memory” to reflect their interpretation as target synaptic weight values. We have also updated the paper’s language to more clearly disambiguate between the “learning rule,” which determines how the memory vector (target synaptic weight vectors) are derived from task variables, and the “plasticity rule,” which governs how these are translated into actual synaptic weight updates. We acknowledge that our manuscript still does not explicitly consider a plasticity rule that is sensitive to continuous error error signals, as our analysis is restricted to binary weights. However, we believe that the updated notation and exposition makes it more clear that our model could be applied in such a case.

Reviewer #1 brought up that our framework cannot capture “single-shot learning, for example, under fear conditioning or if a presented stimulus is astonishing.” Reviewer #2 raised a related question of how our model “relates to the opposite more intuitive idea, that novel surprising experiences should be stored in memory, as the familiar ones are presumably already stored.”

We agree that the built-in inability to consolidate memories after a single experience is a limitation of our model, and that extreme novelty is one factor (among others, such as salience or reward) that might incentivize one-shot consolidation. We have added a comment to the discussion to acknowledge these points (added text in bold): “ Moreover, in real neural circuits, additional factors besides recall, such as reward or salience, are likely to influence consolidation as well. For instance, a sufficiently salient event should be stored in long-term memory even if encountered only once. Furthermore, while in our model familiarity drives consolidation, certain forms of novelty may also incentivize consolidation, raising the prospect of a non-monotonic relationship between consolidation probability and familiarity.” We agree that future work should address the combined influence of recall (as in our model) and other factors on the propensity to consolidate a memory.

Reviewer #1 requested, “a comparison/discussion of the wide range of models on synaptic tagging for consolidation by various types of signals. Notably, studies from Wulfram Gerstner's group (e.g., Brea, J., Clayton, N. S., & Gerstner, W. (2023). Computational models of episodic-like memory in food-caching birds. Nature Communications, 14(1); and studies on surprise).”

We thank the reviewer for the reference, which we have added to the manuscript. The model of Brea et al.(2023) is similar to that of Roxin & Fusi (2013), in that consolidation consists of “copying” synaptic weights from one population to another. As a result, just like the model of Roxin & Fusi (2013), this model does not provide the benefit that our model offers in the context of consolidating repeatedly recurring memories. However, the model of Brea et al. does have other interesting properties – for instance, it affords the ability to decode the age of a memory, which our model does not. We have added a comment on this point in the subsection of the Discussion tilted “Other models of systems consolidation.”

Reviewer #2 noted, “While the article extensively discusses the strengths and advantages of the recall-gated consolidation model, it provides a limited discussion of potential limitations or shortcomings of the model, such as the missing feature of generalization, which is part of previous consolidation models. The model is not compared to other consolidation models in terms of performance and how much it increases the signal-to-noise ratio.”

We agree that our work does not consider the notion of generalization and associated changes to representational geometry that accompany consolidation, which is the focus of many other studies on consolidation. We have further highlighted this limitation in the discussion. Regarding the comparison to other models, this is a tricky point as the desiderata we emphasize in this study (the ability to recall memories that are intermittently reinforced) is not the focus of other studies. Indeed, our focus is primarily on the ability of systems consolidation to be selective in which memories are consolidated, which is somewhat orthogonal to the focus of many other theoretical studies of consolidation. We have updated some wording in the introduction to emphasize this focus.

Additional comments made by reviewer #1

Reviewer #1 pointed out issues in the clarity of Fig. 2A. We have added substantial clarifying text to the figure caption.

Reviewer #1 pointed out lack of clarity in our introduction to the terms “reliability” and “reinforcement.” We have now made it more clear what we mean by these terms the first time they are used.

We have updated our definition of “recall” to use the term “recall factor,” which is how we refer to it subsequently in the paper.

We have made explicit in the main text our simplifying assumption that memories are mean-centered.

We have made consistent our use of “forgetting curve” and “memory trace”.

Additional comments made by reviewer #2

We have added a comment in the discussion acknowledging alternative interpretations of the result of Terada et al. (2021)

We have significantly expanded the discussion of findings about the mushroom body to make it accessible to readers who do not specialize in this area. We hope this clarifies the nature of the experimental finding, which uncovered a circuit that performs a strikingly clean implementation of our model.

The reviewer expresses concern that the songbird study (Tachibana et al., 2022) does not provide direct evidence for consolidation being gated by familiarity of patterns of activity. Indeed, the experimental finding is one-step removed from the direct predictions of our model. That said, the finding – that the rate of consolidation increases with performance – is highly nontrivial, and is predicted by our model when applied to reinforcement learning tasks. We have added a comment to the discussion acknowledging that this experimental support for our model is behavioral and not mechanistic.

We do not regard it as completely trivial that the parallel LTM model performs roughly the same as the STM model, since a slower learning rate can achieve a higher SNR (as in Fig. 2C). Nevertheless we have added wording to the main text around Fig. 4B to note that the result is not too surprising.

We have added a sentence that clarifies the goal / question of our paper earlier on in the introduction.

We have updated Figure 3 by labeling the key components of the schematics and adding more detail to the legend, as suggested by the reviewer. We also reordered the figure panels as suggested.

Additional comments made by reviewer #3:

We have clarified in the main text that Fig. 2C and all results from Fig. 4 onward are derived from an ideal observer model (which we also more clearly define).

We have now emphasized in the main text that the derivations of the recall factors for specific learning rules are derived in the Supplementary Information.

We have highlighted more clearly in the main text that the recall factors associated with specific learning rules may correspond to other notions that do not intuitively correspond to “recall,” and have added a pointer to Fig. 3A where these interpretations are spelled out.

We have added references corresponding to the types of learning rules we consider.

The cutoffs / piecewise-looking behavior of plots in Fig. 4 are primarily the result of finite N, which limits the maximum SNR of the system, rather than coarse sampling of parameter values.

Thank you for pointing out the error in the legend in Fig. 5D (also affected Supp Fig. S7/S8), which is now fixed.

The reference to the nonexistence panel Fig. 5G has been removed.

As the reviewer points out, the use of a binary action output in our reinforcement learning task renders it quite similar to the supervised learning task, making the example less compelling. In the revised manuscript we have updated the RL simulation to use three actions. Note also that in our original submission the network outputs represented action probabilities directly (which is straightforward to do for binary actions, but not for more than two available actions). In order to parameterize a policy when more than two actions are available, we sample actions using a softmax policy, as is more standard in the field and as the reviewer suggested. The associated recall factor is still a product of reward and a “confidence factor,” and the confidence factor is still the value of the network output in the unit corresponding to the chosen action, but in the updated implementation this factor is equal to , similar (though with a sign difference) to the reviewer’s suggestion. We believe these updates make our RL implementation and simulation more compelling, as it allows them to be applied to tasks with arbitrary numbers of actions.

Additional minor comments

The reviewers made a number of other specific line-by-line wording suggestions, typo corrections,

Reviewer #1 (Public Review):

The mechanisms of how axonal projections find their correct target requires the interplay of signalling pathways, and cell adhesion that act over short and long distances. The current study aims to use the small ventral lateral clock neurons (s-LNvs) of the Drosophila clock circuit as a model to study axon projections. These neurons are born during embryonic stages and are part of the core of the clock circuit in the larval brain. Moreover, these neurons are maintained through metamorphosis and become part of the adult clock circuit. The authors use the axon length by means of anti-Pdf antibody or Pdf>GFP as a read-out for the axonal length. Using ablation of the MB- the overall target region of the s-LNvs, the authors find defects in the projections. Next, by using Dscam mutants or knock-down they observe defects in the projections. Manipulations by the DNs - another group of clock neurons - can induce defects in the s-LNvs axonal form, suggesting an active role of these neurons in the morphology of the s-LNvs.

Reviewer #2 (Public Review):

The paper from Liu et al shows a mechanism by which axons can change direction during development. They use the sLNv neurons as a model. They find that the appearance of a new group of neurons (DNs) during post-embryonic proliferation secretes netrins and repels horizontally towards the midline, the axonal tip of the LNvs. The experiments are well done and the results are conclusive.

Author response:

The following is the authors’ response to the original reviews.

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The mechanisms of how axonal projections find their correct target requires the interplay of signalling pathways, and cell adhesion that act over short and long distances. The current study aims to use the small ventral lateral clock neurons (s-LNvs) of the Drosophila clock circuit as a model to study axon projections. These neurons are born during embryonic stages and are part of the core of the clock circuit in the larval brain. Moreover, these neurons are maintained through metamorphosis and become part of the adult clock circuit. The authors use the axon length by means of anti-Pdf antibody or Pdf>GFP as a read-out for the axonal length. Using ablation of the MB- the overall target region of the s-LNvs, the authors find defects in the projections. Next, by using Dscam mutants or knock-down they observe defects in the projections. Manipulations by the DNs - another group of clock neurons- can induce defects in the s-LNvs axonal form, suggesting an active role of these neurons in the morphology of the s-LNvs.

Strengths:

The use of Drosophila genetics and a specific neural type allows targeted manipulations with high precision.

Proposing a new model for a small group of neurons for axonal projections allows us to explore the mechanism with high precision.

Weaknesses:

It is unclear how far the proposed model can be seen as developmental.

The study of changes in fully differentiated and functioning neurons may affect the interpretation of the findings.

We appreciate the reviewer's feedback on the strengths and weaknesses of our study.

We acknowledge the strengths of our research, particularly the precision afforded by using Drosophila genetics and a specific neural type for targeted manipulations, as well as the proposal of a new model for studying axonal projections in a small group of neurons.

We understand the concerns about the developmental aspects of our proposed model and the use of Pdf-GAL4 >GFP as a read-out for the axonal length (revised manuscript Figure 1--figure supplement 1). However, even with the use of Clk856-GAL4 that began to be expressed at the embryonic stage (revised manuscript Figure 3--figure supplement 1) to suppress Dscam expression, the initial segment of the dorsal projection of s-LNvs (the vertical part) remained unaffected. Instead, the projection distance is severely shortened towards the midline, and this defect persists until the adult stage. It is for this reason that we delineate the dorsal projections of s-LNvs into two distinct phases: the vertical and horizontal parts, rather than a mere expansion in correspondence with the development of the larval brain.

Thank you for your valuable feedback, and we have incorporated these considerations into our revised manuscript to enhance the clarity and depth of our research.

Reviewer #2 (Public Review):

Summary:

The paper from Li et al shows a mechanism by which axons can change direction during development. They use the sLNv neurons as a model. They find that the appearance of a new group of neurons (DNs) during post-embryonic proliferation secretes netrins and repels horizontally towards the midline, the axonal tip of the LNvs.

Strengths:

The experiments are well done and the results are conclusive.

Weaknesses:

The novelty of the study is overstated, and the background is understated. Both things need to be revised.

We appreciate your acknowledgment that the experiments were well-executed and the results conclusive. This validation reinforces the robustness of our findings.

We take note of your feedback regarding the novelty of the study being overstated and the background being understated. While axonal projections navigate without distinct landmarks, like the midline or the layers, columns, and segments, they pose more challenges and uncertainties. As highlighted, our key contribution lies in elucidating how axonal projections without clear landmarks are guided, with our research demonstrating how a newly formed cluster of cells at a specific time and location provides the necessary guidance cues for axons.

We value your insights, and we have carefully addressed these points in our manuscript revision to improve the overall quality and presentation of our research.

Recommendations For The Authors:

Reviewer #1 (Recommendations For The Authors):

The overall idea of using the s-LNvs as a model is indeed intriguing. There are genetic tools available to tackle these cells with great precision.

However, based on the stage at which these cells are investigated raises some issues, that I feel are critical to be addressed.

These neurons develop their axonal projections during embryogenesis and are fully functioning when the larvae hatch, thus to investigate axonal pathfinding one would have to address embryonic development.

The larval brain indeed continues to grow during larval life, however extensive work from the Hartenstein lab, Truman lab, and others have shown that the secondary (larval born) neurons do not yet wire into the brain, but stall their axonal projections.

It is thus quite unclear, what the authors are actually studying.

One interpretation could be that the authors observe changes in axon length due to morphological changes in the brain. Indeed, the fact that the MB expands the anatomy of the surrounding neuropil changes too.

Moreover, it is unclear when exactly the Pdf-Gal4 (and other drivers) are active, thus how far (embryonic) development of s-LNvs is affected, or if it's all happening in the differentiated, functioning neuron. (Gal4 temporal delay and dynamics during embryonic development may further complicate the issue). As far as I am aware the MB drivers might already be active during embryonic stages.

Since the raised issue is quite fundamental, I am not sure what might be the best and most productive fashion to address this.

Eg. either to completely re-focus the topic on "neural morphology maintenance" or to study the actual development of these cells.

We thank the reviewer for the detailed and insightful feedback on our study. We have tested whether Pdf-Gal4 could effectively label s-LNv, and tracked the s-LNv projection in the early stage after larvae hatching. We did not observe the PDF antibody staining signal and the GFP signal driven by Pdf-GAL4 when the larvae were newly hatched. At 2-4 hours ALH, PDF signals were primarily concentrated at the end of axons, while GFP signals were mainly concentrated at the cell body. Helfrich-Förster initially detected immunoreactivity for PDF in the brains approximately 4-5 hours ALH. The GFP signal expressed by Pdf-GAL4 driver does have signal delay. However, at 8 hours ALH, the GFP signal strongly co-localized with the PDF signal within the axons (see revised manuscript lines 98-101) (Figure 1—figure supplement 1).

Based on previous research findings and our staining of Clk856-GAL4 >GFP, it is indeed confirmed that the dorsal projection of s-LNvs in Drosophila is formed during the embryonic stage (Figure 3—figure supplement 1). The s-LNvs in first-instar larval Drosophila are capable of detecting signal output and may play a role in regulating certain behaviors. Our selection of tools for characterizing the projection pattern of s-LNv was not optimal, leading us to overlook the crucial detail that the projection had already formed during its embryonic stage.

However, even when employing Clk856-GAL4 to suppress Dscam expression from the embryonic stage, the initial segment of the dorsal projection of s-LNvs (the vertical part) remains unaffected. Instead, the projection distance is severely shortened towards the midline, and this defect persists until the adult stage. It is for this reason that we delineate the dorsal projections of s-LNvs into two distinct phases: the vertical and horizontal parts, rather than a mere expansion in correspondence with the development of the larval brain.

From the results searched in the Virtual Fly Brain (VFB) database (https://www.virtualflybrain.org/), it is clear that the neurons that form synaptic connections with s-LNvs at the adult stage are essentially completely different from the neurons that are associated with them at the L1 larval stage. Thus, most neurons that form synapses with s-LNvs in the early larvae either cease to exist after metamorphosis or assume other roles in the adult stage. Similar to the scenario where Cajal-Retzius cells and GABAergic interneurons establish transient synaptic connections with entorhinal axons and commissural axons, respectively, these cells form a transient circuit with presynaptic targets and subsequently undergo cell death during development. In our model, the neurons that synapse with s-LNvs in early development serve as "placeholders," offering positive or negative cues to guide the axonal targeting of s-LNvs towards their ultimate destination.

Thank you again for your valuable feedback, and we have incorporated these considerations into our revised manuscript to enhance the clarity and depth of our research.

Reviewer #2 (Recommendations For The Authors):

Major:

In the introduction too many revisions are cited and very few actual research papers. This should be corrected and the most significant papers in the field should be cited. For example, there is no reference to the pioneering work from the Christine Holt lab or the first paper looking at axon guidance and guideposts by Klose and Bentley, Isbister et al 1999.

The introduction should encapsulate the actual knowledge based on actual research papers.

We acknowledge your concern regarding the citation of review papers rather than primary research papers in the introduction. Following your suggestion, we have revised the introduction section to incorporate references to relevant research papers.

In the introduction and discussion: The authors cite revisions where the signals that guide axons across different regions including turning are shown and they end up saying: "However, how the axons change their projection direction without well-defined landmarks is still unclear." I think the sentence should be changed. Many things are still not clear but this is not a good phrasing. Maybe they could focus on their temporal finding?

We appreciate the reviewer's feedback and insightful suggestions. We agree that emphasizing the temporal aspect is crucial in our study. However, we also recognize the significance of understanding the origin of signals that guide axonal reorientation at specific locations. While axonal projections navigating without distinct landmarks pose more challenges and uncertainties compared to those guided by prominent landmarks like the midline, our research demonstrates the crucial role of a specific cell population near turning points in providing accurate guidance cues to ensure precise axonal reorientation. We have revised our phrasing in the introduction and discussion to better reflect these key points (see revised manuscript lines 69-71 and 350-354). Thank you for highlighting the significance of focusing on our temporal findings and the complexities involved in studying axonal projection.

Many rather old papers have looked into the effect of repulsive guideposts to guide axon projections. In particular, I can think of the paper from Isbister et al. 1999 (DOI: 10.1242/dev.126.9.2007) that not only shows how semaphoring guides Ti axon projection but also shows how the pattern of expression of sema 2a changes during development to guide the correct projection. I really think that the novelty of the paper should be revised in light of the actual knowledge in the field.

We appreciate the reviewer's reference to the seminal work by Isbister et al. (1999) and the importance of guidepost cells in axon projection guidance, which we have already cited in our revised manuscript. It is crucial to recognize that segmented patterns such as the limb segment traversed by Ti1 neuron projections or neural circuits formed in a layer- or column-specific manner also serve as intrinsic "guideposts," offering valuable insights into axonal pathfinding processes. In our model, explicit guidance cues are lacking. As highlighted, our key contribution lies in elucidating how axonal projections without clear landmarks are guided, with our research demonstrating how a newly formed cluster of cells at a specific time and location provides the necessary guidance cues for axons (see revised manuscript lines 350-354). We have ensured that our revised manuscript reflects these insights and emphasizes the significance of studying axonal guidance in the absence of distinct guideposts. Thank you for underscoring these essential points, which enhance our understanding of axonal projection dynamics.

Minors:

Line 54, the authors start talking about floorplate at the end of a section on Drosophila. Please use “In vertebrates”, or “in invertebrates” or “in Drosophila” etc.. when needed to put things in context.

We thank the reviewer for this suggestion and have modified this sentence. Please refer to lines 62-63 of the revised manuscript.

Line 69: many factors change the axonal outgrowth. The authors are missing the paper from Fernandez et al. 2020, who have shown that unc5 the receptor of netrin induces the stalling for sLNvs projections before the turn. https://doi.org/10.1016/j.cub.2020.04.025

We thank the reviewer for this suggestion and have added this research article. Please refer to line 79 of the revised manuscript.

Line 99: "precisely at the pivotal juncture". It I hard to see how it was done in the figures shown. Can the authors add a small panel with neuronal staining showing this (please no HRP)?

For all figures, tee magenta is too strong and it is really hard to see the sLNvs projections. Can this be sorted, please?

We have depicted the pivotal juncture in the schematic diagram on the left side of Figure 1C. Additionally, we have included a separate column of images without HRP in Figure 1A. Moreover, we have modified the pseudo-color of HRP from magenta to blue to enhance the visualization of the s-LNv projection. The figure legends have also been correspondingly modified.

Line 407: Spatial position relationship between calyx and s-LNvs. OK107-GAL4 labels ... calyx and s-LNvs labeled by, which which.

We have modified it according to your suggestion. Please refer to lines 430-432 of the revised manuscript.

Line 137 typo RPRC

We thank the reviewer for noticing this mistake, which has now been corrected. Please refer to line 148-149 of the revised manuscript.

Section 158-164. the paper from Zhang et al 2019 needs to be cited since they have found the same effect of decreasing Dscam even if they didn't think about horizontal projection.

Thanks to the suggestion, we have included in the manuscript the phenotype observed by Zhang et al. (2019) upon knocking down Dscam1-L in adults. Please refer to lines 170-172 of the revised manuscript.

Line 176: typo senses (instead of sensor).

Thank you for pointing out our mistake. We have modified it according to your suggestion. Please refer to line 189 of the revised manuscript.

Line 193: more than Interesting it is Notable. Add "ubiquitus" knockdown.

Thank you for the suggestion. We have included the word "ubiquitus" to enhance the precision of the narrative. Please refer to line 206 of the revised manuscript.

Line 224: the pattern of expression of the crz cells is not visible where the projections of sLNvs are located. Are they in that region? Or further away?

We've changed the pseudo-color of HRP, and in the updated Figure 5- figure supplement 1, you can see the projection pattern of crz+ cells, positioned close to the end of the s-LNv axon terminal.

Line 243: applied? Do you mean "used"

Thank you for the suggestion. We have revised it at line 256.

Figure 5 Sup1: the schematic shows DNs proliferation that is not visible on the GFP image. Please comment.

We have modified the Figure 5 figure supplementary 1 for 120 h per-GAL4, Pdf-GAL80 >GFP expression pattern. Due to the strong GFP intensity in some DN neurons, there was a loss of GFP signal. Additionally, in Figure 6 figure supplementary 1, we have added co-localization images of DN and s-LNv at 72 h and 96 h. To better illustrate the co-localization information, we have shown only a portion of the layers in the right panel. We hope these additions clarify your concerns.

Line 251: cite Fernandez et al. 2020 with Purohit et al 2012.

We have modified it according to your suggestion. Please refer to line 264 of the revised manuscript.

Line 272: you have not shown synergistic effects because you have not modulated both pathways at the same time. You should talk about complementary.

We have modified it according to your suggestion at lines 25, 285, 439.

Author response:

The following is the authors’ response to the original reviews.

Reviewer #1:

(1) Point for more elaborate discussion: Apparently the timescale of negative feedback signals is conserved between endothelial cell migration in vitro (with human cells) and endothelial migration during the formation of ISVs in zebrafish. What do you think might be an explanation for such conserved timescales? Are there certain processes within cytoskeletal tension build up that require this quantity of time to establish? Or does it relate to the time that is needed to begin to express the YAP/TAZ target genes that mediate feedback?

The underlying mechanisms responsible for the conserved timescale is a major direction that we continue to explore. Localization of YAP/TAZ to the nucleus is likely not rate-limiting. We showed previously that acute RhoA activation produced significant YAP/TAZ nuclear localization within minutes, while subsequent co-transcriptional activity aligned with the gene expression dynamics observed here (Berlew et al., 2021). We hypothesize that the dynamics of YAP/TAZdependent transcription and the translation of those target genes are rate-limiting for initial feedback loop completion (tic = 4 hours). This is supported by work from us and others in a variety of cell lines showing YAP/TAZ transcriptional responses take place during the first few hours after activation. (Franklin et al., 2020; Mason et al., 2019; Plouffe et al., 2018) While our data identify mediators of initial feedback loop completion, the molecular effectors that determine the timescale of new cytoskeletal equilibrium establishment (teq = 8 hours) remain unclear.

(2) Do you expect different timescales for slower endothelial migratory processes (e.g. for instance during fin vascular regeneration which takes days)?

We selected the ISV development model because it exhibits similar migratory kinetics to our previously-explored human ECFC migration in vitro. The comparable kinetics allowed us to study dynamics of the feedback loop in vivo on similar time scales, but we have not explored models featuring either slower or faster dynamics. 

It would be interesting to test how feedback dynamics are impacted in distinct endothelial migratory processes. Our data suggest that the feedback loop is necessary for persistent migration; however, YAP and TAZ respond to a diversity of upstream regulators in addition to mechanical signals, which might depend on the process of vascular morphogenesis. For example, after fin amputation, inflammation and tissue regeneration may impact the biochemical and mechanical environment experienced by the endothelium. Additionally, cells display different migratory behaviors in ISV morphogenesis compared to fin regeneration. During ISV formation, sprouting tip cells migrate dorsally through avascular tissue, followed by stalk cells. (Ellertsdóttir et al., 2010) In contrast, the fin vasculature regenerates by forming an intermediate vascular plexus, where some venous-derived endothelial cells migrate towards the sprouting front, while others migrate against it. (Xu et al., 2014) We are excited to study the role of this feedback loop in these different modes of neovessel formation in future studies.

(3) Is the ~4hrs and 8hrs feedback time window a general property or does it differ between specific endothelial cell types? In the veins the endothelial cells generate less stress fibers and adhesions compared to in the arteries. Does this mean that there might be a difference in the feedback time window, or does that mean that certain endothelial cell types may not have such YAP/TAZcontrolled feedback system?

Recent studies suggest that venous endothelial cells are the primary endothelial subtype responsible for blood vessel morphogenesis. (Lee et al., 2022, 2021; Xu et al., 2014) They are highly motile and mechanosensitive, migrating against blood flow. (Lee et al., 2022) The Huveneers group has shown that the actin cytoskeleton is differently organized in adult arteries and veins in response to biomechanical properties of its extracellular matrix, rather than intrinsic differences between arterial and venous cells. (van Geemen et al., 2014) This suggests that arterial and venous cells have distinct cytoskeletal setpoints due to mechanical cues in their environment (Price et al., 2021). We expect this to impact the degree of cytoskeletal remodeling and cell migration at equilibrium, rather than the kinetics of the feedback loop per se, though we have not yet tested this hypothesis. Testing these predictions on cytoskeletal setpoint stability and adaptation is a major direction that we continue to explore. 

(4) The experiments are based on perturbations to prove that transcriptional feedback is needed for endothelial migration. What would happen if the feedback systems is always switched on? An experiment to add might be to analyse the responsiveness of endothelial cells expressing constitutively active YAP/TAZ.

This is a problem that we are actively pursuing. Though the feedback system forms a coherent loop, we anticipate that the identity of the node of the loop selected for constitutive activation will influence the outcome, depending on whether that node is rate-limiting for feedback kinetics and the extent of intersection of that node with other signaling events in the cell. For example, we have observed that constitutive YAP activation drives profound changes to the transcriptional landscape including, but not limited to, RhoA signaling (Jones et al., 2023). We further anticipate that constitutive activation of feedback loop nodes may alter feedback dynamics, while dynamic or acute perturbation will be required to dissect these contributions in real time. For these reasons, ongoing work in the lab is pursuing these questions using optogenetic tools that enable precise spatial and temporal control (Berlew et al., 2021).   

(5) To investigate the role of YAP-mediated transcription in an accurate time-dependent manner the authors may consider using the recently developed optogenetic YAP translocation tool: https://doi.org/10.15252/embr.202154401

We are enthusiastic about the power of optogenetics to interrogate the nodes and timescales of this feedback system, and we are now funded to pursue this line of research. 

Reviewer #2:

The idea is intriguing, but it is not clear how the feedback actually works, so it is difficult to determine if the events needed could occur within 4 hrs. Specifically, it is not clear what gene changes initiated by YAP/TAZ translocation eventually lead to changes in Rho signaling and contractility. Much of the evidence to support the model is preliminary. Some of the data is consistent with the model, but alternative explanations of the data are not excluded. The fish washout data is quite interesting and does support the model. It is unclear how some of the in vitro data supports the model and excludes alternatives.

Major strengths:

The combination of in vitro and in vivo assessment provides evidence for timing in physiologically relevant contexts, and a rigorous quantification of outputs is provided. The idea of defining temporal aspects of the system is quite interesting.

Major weaknesses:

The evidence for a "loop" is not strong; rather, most of the data can also be interpreted as a linear increase in effect with time once a threshold is reached. Washout experiments are key to setting up a time window, yet these experiments are presented only for the fish model. A major technical challenge is that siRNA experiments take time to achieve depletion status, making precise timing of events on short time scales problematic. Also, Actinomycin D blocks most transcription so exposure for hours likely leads to secondary and tertiary effects and perhaps effects on viability. No RNA profiling is presented to validate proposed transcriptional changes.

We thank the reviewer for these helpful suggestions. We have expanded our explanation of the history and known mediators of the feedback loop in the introduction. We and, independently, the Huveneers group recently reported that human endothelial cells maintain cytoskeletal equilibrium for persistent motility through a YAP/TAZ-mediated feedback loop that modulates cytoskeletal tension. (Mason et al., 2019; van der Stoel et al., 2020) Because YAP and TAZ are activated by tension of the cytoskeleton (Dupont et al., 2011), suppression of cytoskeletal tension by YAP/TAZ transcriptional target genes constitutes a negative feedback loop (Fig. 1A). We described key components of this cell-intrinsic feedback loop, which acts as a control system to maintain cytoskeletal homeostasis for persistent motility via modulation of Rho-ROCK-myosin II activity. (Mason et al., 2019) Both we and the Huveneers group found that perturbation of genes and pathways regulated by YAP/TAZ mechanoactivation can functionally rescue motility in YAP/TAZ-depleted cells (e.g., RhoA/ROCK/myosin II, NUAK2, DLC1). (Mason et al., 2019; van der Stoel et al., 2020) We further showed previously that both YAP/TAZ depletion and acute YAP/TAZ-TEAD inhibition consistently increased stress fiber and FA maturation and arrested cell motility, accounting for these limitations of siRNA. (Mason et al., 2019)

Enduring limitations to the temporal, spatial, and cell-specific control of the genetic and pharmacologic methods have inspired us to initiate alternative approaches, which are the subject of ongoing efforts. Further research will be necessary in the zebrafish to determine the extent to which the observed migratory dynamics are driven by cytoskeletal arrest. 

To identify early YAP/TAZ-regulated transcriptional changes, we have added RNA profiling of control and YAP/TAZ depleted cells cultured on stiff matrices for four hours. Genes upregulated by YAP/TAZ depletion were enriched for Gene Ontology (GO) terms associated with Rho protein signal transduction, vascular development, cellular response to vascular endothelial growth factor (VEGF) stimulus, and endothelial cell migration (Fig. 9B). These data support a role for YAP and TAZ as negative feedback mediators that maintain cytoskeletal homeostasis for endothelial cell migration and vascular morphogenesis.  

Reviewer #3:

The authors used ECFC - endothelial colony forming cells (circulating endothelial cells that activate in response to vascular injury).

Q: Did the authors characterize these cells and made sure that they are truly endothelial cells - for example examine specific endothelial markers, arterial-venous identity markers & Notch signalling status, overall morphology etc prior to the start of the experiment. How were ECFC isolated from human individuals, are these "healthy" volunteers - any underlying CVD risk factors, cells from one patient or from pooled samples, what injury where these humans exposed to trigger the release of the ECPFs into the circulation, etc. The materials & methods on ECFC should be expanded.

Human umbilical cord blood-derived ECFCs were isolated at Indiana University School of Medicine and kindly provided by Dr Mervin Yoder. Cells were cultured as described by the Yoder group (Rapp et al., 2011) and our prior paper (Mason et al., 2019). We have expanded the materials and methods section to describe the source and characterization of these cells.

The authors suggest that loss of YAP/TAZ phenocopies actinomycin-D inhibition - "both transcription inhibition and YAP/TAZ depletion impaired polarization, and induced robust ventral stress fiber formation and peripheral focal adhesion maturation". However, the cell size of actinomycin-D treated cells (Fig. 1B, top right panel), differs from the endothelial cell size upon siYAP/TAZ (Fig. 1E, top right panel) - and vinculin staining seems more pronounced in actinomycin-D treated cells (B, bottom right) when compared to siYAP/TAZ group. Cell shape is defined by acto-myosin tension.

Q: Besides Fraction of focal adhesion >1um; focal adhesion number did the authors measure additional parameters related to cytoskeleton remodelling / focal adhesions that can substantiate their statement on similarity between loss of YAP/TAZ and actinomycin-D treatment. Would it be possible to make a more specific genetic intervention (besides YAP/TAZ) interfering with the focal adhesion pathway as opposed to the broad spectrum inhibitor actinomyocin-D.

Our previous paper (Mason et al., 2019) delineated the mechanistic relationships between YAP/TAZ signaling, focal adhesion turnover, actomyosin polymerization, and the intervening mechanisms of myosin regulation. Specifically, we demonstrated that YAP/TAZ regulate the myosin phosphatase kinase, NUAK2, and ARHGAP genes to mediate this feedback. Expanding on this work, the current study aimed to define the temporal kinetics of the cytoskeletal mechanotransductive feedback in vitro and in vivo. We used actinomycin-D and YAP/TAZ depletion to interrogate the role of transcriptional regulation and YAP/TAZ signaling, respectively. In this revision, we have added RNA profiling that identifies early YAP/TAZ-regulated transcriptional changes and further points to other molecular mediators of focal adhesions (e.g. TRIO, RHOB, THBS1) that will be the subjects of future studies.    

Q: Does the actinomycin-D treatment affect responsiveness to Vegf? induce apoptosis or reduce survival of the ECFC?

We have not looked specifically at the effect of actinomycin-D treatment on responsiveness to VEGF. However, actinomycin-D has been reported to reduce transcription of VEGF receptors (E et al., 2012). In contrast, we found that YAP/TAZ depletion upregulated GO terms associated with endothelial cell migration and response to VEGF stimulus (Fig. 9B), as well as receptors to angiogenic growth factors, including KDR and FLT4 (Fig. 9E). These results suggest YAP/TAZ depleted cells may be more sensitive to VEGF stimulation but remain nonmotile due to cytoskeletal arrest.

We showed previously that long-term treatment with actinomycin-D reduces ECFC survival (Mason et al., 2019).

Q: Which mechanism links ECM stiffness with endothelial surface area in the authors scenario. In zebrafish, activity of endothelial guanine exchange factor Trio specifically at endothelial cell junctions (Klems, Nat Comms, 2020) and endoglin in response to hemodynamic factors (Siekmann, Nat Cell Biol 2017) have been show to control EC shape/surface area - do these factors play a role in the scenario proposed by the authors.

Our new transcriptional profiling indicates both Trio and endoglin are regulated through YAP and TAZ in human ECFCs. We plan to follow up on these findings.

Q: The authors report that EC migrate faster on stiff substrate, and concomitantly these cells have a larger surface area. What is the physiological rationale behind these observations. Did the authors observe such behaviors in their zebrafish ISV model? How do these observations integrate with the tip - stalk cell shuffling model (Jakobsson & Gerhardt, Nat Cell Biol, 2011) and Notch activity in developing ISVs.

This question raises important distinctions between the mode of migration in ISV morphogenesis and endothelial cells adherent to substrates. Cells behave and respond to mechanical cues differently in 2D vs. 3D matrices. (LaValley and Reinhart-King, 2014) Additionally, the microenvironment in vivo is much more complex, combining numerous biochemical signals and changing mechanical properties. (Whisler et al., 2023) We are actively investigating the downstream targets of YAP/TAZ mechanotransduction and how that integrates with other pathways known to regulate vascular morphogenesis, such as Notch signaling. 

The authors examined the formation of arterial intersegmental vessels in the trunk of developing zebrafish embryos in vivo. They used a variety of pharmacological inhibitors of transcription and acto-myosin remodelling and linked the observed morphological changes in ISV morphogenesis with changes in endothelial cell motility.

Q: Reduced formation and dorsal extension of ISVs may have several reasons, including reduced EC migration and proliferation. The Tg(fl i1a:EGFP) reporter however is not the most suitable line to monitor migration of individual endothelial cells. Can the authors repeat the experiments in Tg(fl i1a:nEGFP); Tg(kdrl:HRAS-mCherry) double transgenics to visualize movement-migration of the individual endothelial cells and EC proliferation events, in the different treatment regimes.

So far, we have not tracked individual endothelial cells during ISV morphogenesis. We agree this is the best approach and are pursuing a similar technique for these experiments.

ISV formation is furthermore affected by Notch signalling status and a series of (repulsive) guidance cues.

Q: Does de novo blockade of gene expression with Actinomycin D affect Notch signalling status, expression of PlexinD - sFlt1, netrin1 or arterial-venous identify genes.

While we have not performed gene expression analysis under the Actinomycin D condition, Actinomycin D functions as a broad transcription inhibitor. We are currently pursuing the downstream targets of YAP/TAZ mechanotransduction in both ECFCs and zebrafish.

Remark: The authors may want to consider using the Tg(fl i1:LIFEACT-GFP) reporter for in vivo imaging of actin remodelling events.

We thank the reviewer for their helpful suggestion.

Remark: the authors report "As with broad transcription inhibition, in situ depletion of YAP and TAZ by RNAi arrested cell motility, illustrated here by live-migration sparklines over 10 hours: siControl: , siYAP/TAZ: (25 μm scale-bar: -)". Can the authors make a separate figure panel for this, how many cells were measured?

Please refer to our previous publication for the complete details on this data (Mason et al., 2019). We have added the citation in the text.

Remark: in the wash-out experiments, exposure to the inhibitors is not the same in the different scenarios - could it be that the longer exposure time induces "toxic" side effect that cannot be "washed out" when compared to the short treatment regimes?

This is a possible limitation of the pharmacological approach and have included it in the discussion section. We are currently exploring alternative approaches to interrogate the timescale of the feedback loop more precisely.  

References

Berlew EE, Kuznetsov IA, Yamada K, Bugaj LJ, Boerckel JD, Chow BY. 2021. Single-Component Optogenetic Tools for Inducible RhoA GTPase Signaling. Advanced Biology 5:2100810. doi:10.1002/adbi.202100810

Dupont S, Morsut L, Aragona M, Enzo E, Giulitti S, Cordenonsi M, Zanconato F, Le Digabel J,Forcato M, Bicciato S, Elvassore N, Piccolo S. 2011. Role of YAP/TAZ in mechanotransduction. Nature 474:179–183. doi:10.1038/nature10137

E G, Cao Y, Bhattacharya S, Dutta S, Wang E, Mukhopadhyay D. 2012. Endogenous Vascular Endothelial Growth Factor-A (VEGF-A) Maintains Endothelial Cell Homeostasis by Regulating VEGF Receptor-2 Transcription. J Biol Chem 287:3029–3041. doi:10.1074/jbc.M111.293985

Ellertsdóttir E, Lenard A, Blum Y, Krudewig A, Herwig L, Affolter M, Belting H-G. 2010. Vascular morphogenesis in the zebrafish embryo. Developmental Biology, Special Section: Morphogenesis 341:56–65. doi:10.1016/j.ydbio.2009.10.035

Franklin JM, Ghosh RP, Shi Q, Reddick MP, Liphardt JT. 2020. Concerted localization-resets precede YAP-dependent transcription. Nat Commun 11:4581. doi:10.1038/s41467-02018368-x

Jones DL, Hallström GF, Jiang X, Locke RC, Evans MK, Bonnevie ED, Srikumar A, Leahy TP, Nijsure MP, Boerckel JD, Mauck RL, Dyment NA. 2023. Mechanoepigenetic regulation of extracellular matrix homeostasis via Yap and Taz. Proceedings of the National Academy of Sciences 120:e2211947120. doi:10.1073/pnas.2211947120

LaValley DJ, Reinhart-King CA. 2014. Matrix stiffening in the formation of blood vessels. Advances in Regenerative Biology 1:25247. doi:10.3402/arb.v1.25247

Lee H-W, Shin JH, Simons M. 2022. Flow goes forward and cells step backward: endothelial migration. Exp Mol Med 54:711–719. doi:10.1038/s12276-022-00785-1

Lee H-W, Xu Y, He L, Choi W, Gonzalez D, Jin S-W, Simons M. 2021. Role of Venous Endothelial Cells in Developmental and Pathologic Angiogenesis. Circulation 144:1308–1322. doi:10.1161/CIRCULATIONAHA.121.054071

Mason DE, Collins JM, Dawahare JH, Nguyen TD, Lin Y, Voytik-Harbin SL, Zorlutuna P, Yoder MC, Boerckel JD. 2019. YAP and TAZ limit cytoskeletal and focal adhesion maturation to enable persistent cell motility. Journal of Cell Biology 218:1369–1389. doi:10.1083/jcb.201806065

Plouffe SW, Lin KC, Moore JL, Tan FE, Ma S, Ye Z, Qiu Y, Ren B, Guan K-L. 2018. The Hippo pathway effector proteins YAP and TAZ have both distinct and overlapping functions in the cell. J Biol Chem 293:11230–11240. doi:10.1074/jbc.RA118.002715

Price CC, Mathur J, Boerckel JD, Pathak A, Shenoy VB. 2021. Dynamic self-reinforcement of gene expression determines acquisition of cellular mechanical memory. Biophysical Journal 120:5074–5089. doi:10.1016/j.bpj.2021.10.006

Rapp BM, Saadatzedeh MR, Ofstein RH, Bhavsar JR, Tempel ZS, Moreno O, Morone P, Booth DA, Traktuev DO, Dalsing MC, Ingram DA, Yoder MC, March KL, Murphy MP. 2011. Resident Endothelial Progenitor Cells From Human Placenta Have Greater Vasculogenic Potential Than Circulating Endothelial Progenitor Cells From Umbilical Cord Blood. Cell Med 2:85–96. doi:10.3727/215517911X617888

Tammela T, Zarkada G, Nurmi H, Jakobsson L, Heinolainen K, Tvorogov D, Zheng W, Franco CA, Murtomäki A, Aranda E, Miura N, Ylä-Herttuala S, Fruttiger M, Mäkinen T, Eichmann A, Pollard JW, Gerhardt H, Alitalo K. 2011. VEGFR-3 controls tip to stalk conversion at vessel fusion sites by reinforcing Notch signalling. Nat Cell Biol 13:1202–1213. doi:10.1038/ncb2331

van der Stoel M, Schimmel L, Nawaz K, van Stalborch A-M, de Haan A, Klaus-Bergmann A, Valent ET, Koenis DS, van Nieuw Amerongen GP, de Vries CJ, de Waard V, Gloerich M, van Buul JD, Huveneers S. 2020. DLC1 is a direct target of activated YAP/TAZ that drives collective migration and sprouting angiogenesis. Journal of Cell Science 133:jcs239947. doi:10.1242/jcs.239947

van Geemen D, Smeets MWJ, van Stalborch A-MD, Woerdeman LAE, Daemen MJAP, Hordijk PL, Huveneers S. 2014. F-Actin–Anchored Focal Adhesions Distinguish Endothelial Phenotypes of Human Arteries and Veins. Arteriosclerosis, Thrombosis, and Vascular Biology 34:2059–2067. doi:10.1161/ATVBAHA.114.304180

Whisler J, Shahreza S, Schlegelmilch K, Ege N, Javanmardi Y, Malandrino A, Agrawal A, Fantin A, Serwinski B, Azizgolshani H, Park C, Shone V, Demuren OO, Del Rosario A, Butty VL, Holroyd N, Domart M-C, Hooper S, Szita N, Boyer LA, Walker-Samuel S, Djordjevic B, Sheridan GK, Collinson L, Calvo F, Ruhrberg C, Sahai E, Kamm R, Moeendarbary E. 2023. Emergent mechanical control of vascular morphogenesis. Science Advances 9:eadg9781. doi:10.1126/sciadv.adg9781

Xu C, Hasan SS, Schmidt I, Rocha SF, Pitulescu ME, Bussmann J, Meyen D, Raz E, Adams RH, Siekmann AF. 2014. Arteries are formed by vein-derived endothelial tip cells. Nat Commun 5:5758. doi:10.1038/ncomms6758

eLife assessment

This valuable manuscript delineates the role of YAP/TAZ-dependent transcriptional suppression in a mechanodransductive feedback loop. The evidence presented in the manuscript is generally solid. However, compared to an earlier version, some concerns remain. In particular, the in vivo validation should be strengthened, and the in vitro and in vivo models used in this work should be carefully compared in order to improve the main message of the manuscript.

Reviewer #1 (Public Review):

This manuscript puts forward the concept that there is a specific time window during which YAP/TAZ driven transcription provides feedback for optimal endothelial cell adhesion, cytoskeletal organization and migration. The study follows up on previous elegant findings from this group and others which established the importance of YAP/TAZ-mediated transcription for persistent endothelial cell migration. The data presented here extends the concept at two levels: first, the data may explain why there are differences between experimental setups where YAP/TAZ activity are inhibited for prolonged times (e.g. cultures of YAP knockdown cells), versus experiments in which the transient inhibition of YAP/TAZ and (global) transcription affects endothelial cell dynamics prior to their equilibrium.

All experiments are convincing, clearly visualized and quantified.

The strength of the paper is that it clearly indicates that there are temporal controlled feedback systems which which is important for endothelial collective cell behavior.

A limitation of the study is that the inhibitory studies in vivo may include off-target effects as well. Future endeavors, including specific knockout models, optogenetics and/or transgenic zebrafish lines that visualize endothelial cell properties (proliferation and migration) will be informative to track individual endothelial cell responses upon feedback signals.

Reviewer #2 (Public Review):

Summary:

Here the effect of overall transcription blockade, and then specifically depletion of YAP/TAZ transcription factors was tested on cytoskeletal responses, starting from a previous paper showing YAP/TAZ-mediated effects on the cytoskeleton and cell behaviors. Here, primary endothelial cells were assessed on substrates of different stiffness and parameters such as migration, cell spreading, and focal adhesion number/length were tested upon transcriptional manipulation. Zebrafish subjected to similar manipulations were also assessed during the phase of intersegmental vessel elongation. The conclusion was that there is a feedback loop of 4 hours that is important for the effects of mechanical changes to be translated into transcriptional changes that then permanently affect the cytoskeleton.

The idea is intriguing and a previous paper contains data supporting the overall model. The fish washout data is quite interesting and supports the kinetics conclusions. New transcriptional profiling in this version supports that cytoskeletal genes are differentially regulated with YAP/TAZ manipulations.

Major strengths: The combination of in vitro and in vivo assessment provides evidence for timing in physiologically relevant contexts, and rigorous quantification of outputs is provided. The idea of defining temporal aspects of the system is quite interesting. New RNA profiling supports the model.

Weaknesses:

Actinomycin D blocks most transcription so exposure for hours likely leads to secondary and tertiary effects and perhaps effects on viability.

Reviewer #4 (Public Review):

Summary:

Mason DE et al. have extended their previous study on continuous migration of cells regulated by a feedback loop that controls gene expression by YAP and TAZ. Time scale of the negative feedback loop is derived from the authors' adhesion-spreading-polarization-migration (ASPM) assay. Involvement of transcription-translation in the negative feedback loop is evidenced by the experiments using Actinomycin D. The time scale of mechanotransduction-dependent feedback demonstrated by cytoskeletal alteration in the actinomycin D-treated endothelial colony forming cells (ECFCs) and that shown in the ECFCs depleted of YAP/TAZ by siRNA. The authors examine the time scale when ECFCs are attached to MeHA matrics (soft, moderate, and stiff substrate) and show the conserved time scale among the conditions they use, although instantaneous migration, cell area, and circularity vary. Finally, they tried to confirm that the time scale of the feedback loop-dependent endothelial migration by the effect of washout of Actinomycin D (inhibition of gene transcription), Puromycin (translational inhibition), and Verteporfin (YAP/TAZ inhibitor) on ISV extension during sprouting angiogenesis. They conclude that endothelial motility required for vascular morphogenesis is regulated by mechanotransduction-mediated feedback loop that is dependent on YAP/TAZ-dependent transcriptional regulation.

Strengths:

The authors conduct ASPM assay to find the time scale of feedback when ECFCs attach to three different matrics. They observe the common time scale of feedback. Thus, under very specific conditions they use, the reproducibility is validated by their ASPM assay. The feedback loop mediated by inhibition of gene expression by Actinomycin D is similar to that obtained from YAP/TAZ-depleted cells, suggesting the mechanotranduction might be involved in the feedback loop. The time scale representing infection point might be interesting when considering the continuous motility in cultured endothelial cells, although it might not account for the migration of endothelial cells that is controlled by a wide variety of extracellular cues. In vivo, stiffness of extracellular matrix is merely one of the cues.

Weaknesses:

ASPM assay is based on attachment-dependent phenomenon. The time scale including the inflection point determined by ASPM experiments using cultured cells and the mechanotransduction-based theory do not seem to fit in vivo ISV elongation. Although it is challenging to find the conserved theory of continuous cell motility of endothelial cells, the data is preliminary and does not support the authors' claim. There is no evidence that mechanotransduction solely determines the feedback loop during elongation of ISVs. The points to be addressed are listed in recommendations for the authors.

Please also add 'the Trait Selector BrAPP":

Valentin wrote the section on the BrAPI Mapper project and has made significant contributions to the BrAPI compatibility of the BrAPI Mapper, MGIS projects and the Trait Selector BrAPP.

Version 3 of this preprint has been peer-reviewed and recommended by Peer Community in Genomics.<br /> See the peer reviews and the recommendation.

Reviewer #3 (Public Review):

Summary:

In Okholm et al., the authors evaluate the functional impact of circHIPK3 in bladder cancer cells. By knocking it down and performing an RNA-seq analysis, the authors found a thousand deregulated genes which look unaffected by miRNAs sponging function and that are, instead, enriched for a 11-mer motif. Further investigations showed that the 11-mer motif is shared with the circHIPK3 and able to bind the IGF2BP2 protein. The authors validated the binding of IGF2BP2 and demonstrated that IGF2BP2 KD antagonizes the effect of circHIPK3 KD and leads to the upregulation of genes containing the 11-mer. Among the genes affected by circHIPK3 KD and IGF2BP2 KD, resulting in downregulation and upregulation respectively, the authors found STAT3 gene which also consistently leads to the concomitant upregulation of one of its targets TP53. The authors propose a mechanism of competition between circHIPK3 and IGF2BP2 triggered by IGF2BP2 nucleation, potentially via phase separation.

Strengths:

The number of circRNAs continues to drastically grow however the field lacks detailed molecular investigations. The presented work critically addresses some of the major pitfalls in the field of circRNAs and there has been a careful analysis of aspects frequently poorly investigated. The time-point KD followed by RNA-seq, investigation of miRNAs-sponge function of circHIPK3, identification of 11-mer motif, identification and validation of IGF2BP2, and the analysis of copy number ratio between circHIPK3 and IGF2BP2 in assessing the potential ceRNA mode of action has been extensively explored and, comprehensively convincing.

Weaknesses:

The authors addressed the majority of the weak points raised initially. However, the role played by the circHIPK3 in cancer remains elusive and not elucidated in full in this study.

Overall, the presented study surely adds some further knowledge in describing circHIPK3 function, its capability to regulate some downstream genes, and its interaction and competition for IGF2BP2. However, whereas the experimental part sounds technically logical, it remains unclear the overall goal of this study and the achieved final conclusions.

This study is a promising step forward in the comprehension of the functional role of circHIPK3. These data could possibly help to better understand the circHIPK3 role in cancer.

eLife assessment

This work explores the role of one the most abundant circRNAs, circHIPK3, in bladder cancer cells, showing with convincing data that circHIPK3 depletion affects thousands of genes and that those downregulated (including STAT3) share an 11-mer motif with circHIPK3, corresponding to a binding site for IGF2BP2. The experiments demonstrate that circHIPK3 can compete with the downregulated mRNAs targets for IGF2BP2 binding and that IGF2BP2 depletion antagonizes the effect of circHIPK3 depletion by upregulating the genes containing the 11-mer. These important findings contribute to the growing recognition of the complexity of cancer signaling regulation and highlight the intricate interplay between circRNAs and protein-coding genes in tumorigenesis.

Reviewer #1 (Public Review):

In this work the authors propose a new regulatory role for one of the most abundant circRNAs, circHIPK3. They demonstrate that circHIPK3 interacts with an RNA binding protein (IGF2BP2), sequestering it away from its target mRNAs. This interaction is shown to regulate the expression of hundreds of genes that share a specific sequence motif (11-mer motif) in their untranslated regions (3'-UTR), identical to one present in circHIPK3 where IGF2BP2 binds. The study further focuses on the specific case of STAT3 gene, whose mRNA product is found to be downregulated upon circHIPK3 depletion. This suggests that circHIPK3 sequesters IGF2BP2, preventing it from binding to and destabilizing STAT3 mRNA. The study presents evidence supporting this mechanism and discusses its potential role in tumor cell progression. These findings contribute to the growing complexity of understanding cancer regulation and highlight the intricate interplay between circRNAs and protein-coding genes in tumorigenesis.

Strengths:

The authors show mechanistic insight into a proposed novel "sponging" function of circHIPK3 which is not mediated by sequestering miRNAs but rather a specific RNA binding protein (IGF2BP2). They address the stoichiometry of the molecules involved in the interaction, which is a critical aspect that is frequently overlooked in this type of study. They provide both genome-wide analysis and a specific case (STAT3) which is relevant for cancer progression. Overall, the authors have significantly improved their manuscript in their revised version.

Weaknesses:

There are seemingly contradictory effects of circHIPK3 and STAT3 depletion in cancer progression. However, the authors have addressed these issues in their revised manuscript, incorporating potential reasons that might explain such complexity.

Reviewer #2 (Public Review):

The manuscript by Okholm and colleagues identified an interesting new instance of ceRNA involving a circular RNA. The data are clearly presented and support the conclusions. Quantification of the copy number of circRNA and quantification of the protein were performed, and this is important to support the ceRNA mechanism.

This is the second rebuttal and the authors further improved the manuscript. The data are of interest to the large spectrum of readers of the journal.

Comments on revision:

The authors explain that they have compared primer efficiencies of two linear Laccase version amplicons and their divergent primers targeting circHIPK3 using amplification standard curves (not shown). They claim that all amplicons were found to be directly comparable, ensuring that their estimation of cirRNA:lineal ratio estimation by RT-qPCR was accurate. I agree that this is not a technically trivial experiment. However, for this measurement to be valid, it is not enough to compare the efficiencies of primers using cDNA/DNA standard curves in the context of the qPCR reaction alone. Instead, one should perform the full RT-qPCR tandem reactions in the context of standard curves of the specific RNAs (for example, obtained by in vitro synthesis). RNA absolute amounts in these standard curves should be known in order to compare the different RNA species (linear or circular).

I do not have major concerns about this issue.

Author response:

The following is the authors’ response to the previous reviews.

Reviewer #1 (Recommendations For The Authors): 

Major points about revised manuscript 

(1) While I acknowledge that the Laccase2 vector is probably the best available in terms of its clean circRNA-expression potential, the authors still lack an estimation of the circRNA overexpression efficiency, specifically the circular-to-linear expression ratio. In their second rebuttal letter, the authors argue that they do not have the option to use another probe and that they are limited by the Backsplicing junction (BSJ)-specific one. I assume they mean that such a probe might only partially hybridize with the linear form and therefore give a poor or no signal in the Northern blot. However, in this referee's opinion, it is precisely because of this limitation that the authors should have used another probe against both the linear and circular RNAs to simultaneously and quantitatively detect both isoforms. This would have allowed them to reliably estimate a circular-to-linear ratio. Perhaps the linear isoform is indeed not expressed or is very low for this circRNA overexpression vector, but the probe used by the authors does not prove it. I think that this addition to the manuscript is not strictly necessary at this stage, but it would certainly improve the results.  

We fully agree with this recommendation. Our efforts to show this using northern blotting was unfortunately unsuccesful due to background signal. To accommodate the question about circ-to-linear ratio, we instead used an RT-qPCR strategy to measure the linear vs circRNA expression derived from the LaccasecircHIPK3 expression constructs/cell lines. To be able to compare obtained results from different amplicons, we measured primer efficiencies (using amplification standard curves – not shown) of two linear Laccase version amplicons and our divergent primers targeting circHIPK3, which were found to be directly comparable. Using these primer sets in RT-qPCR on the same RNA preparation (total cellular RNA) from the northern blot (Supplementary figure S5H) revealed a ~4 fold higher expression of circHIPK3 compared to linear precursor RNA (Supplementary Figure S5I). 

This demonstrates that the Laccase vector system efficiently produces circHIPK3 RNA as expected. 

The few changes to the manuscript (results section text and reference to Supplementary Figure S5I) has been highlighted in yellow. The materials and methods section and Table S1 have been modified to include description of RTqPCR and specific primers.

Reviewer #1 (Public Review):

Plasticity in the basolateral amygdala (BLA) is thought to underlie the formation of associative memories between neutral and aversive stimuli, i.e. fear memory. Concomitantly, fear learning modifies the expression of BLA theta rhythms, which may be supported by local interneurons. Several of these interneuron subtypes, PV+, SOM+, and VIP+, have been implicated in the acquisition of fear memory. However, it was unclear how they might act synergistically to produce BLA rhythms that structure the spiking of principal neurons so as to promote plasticity. Cattani et al. explored this question using small network models of biophysically detailed interneurons and principal neurons.

Using this approach, the authors had four principal findings:<br /> (1) Intrinsic conductances in VIP+ interneurons generate a slow theta rhythm that periodically inhibits PV+ and SOM+ interneurons, while disinhibiting principal neurons.<br /> (2) A gamma rhythm arising from the interaction between PV+ and principal neurons establishes the precise timing needed for spike-timing-dependent plasticity.<br /> (3) Removal of any of the interneuron subtypes abolishes conditioning-related plasticity.<br /> (4) Learning-related changes in principal cell connectivity enhance the expression of slow theta in the local field potential.

The strength of this work is that it explores the role of multiple interneuron subtypes in the formation of associative plasticity in the basolateral amygdala. The authors use biophysically detailed cell models that capture many of their core electrophysiological features, which helps translate their results into concrete hypotheses that can be tested in vivo. Moreover, they try to align the connectivity and afferent drive of their model with those found experimentally. However, the weakness is that their attempt to align with the experimental literature (specifically Krabbe et al. 2019) is performed inconsistently. Some connections between cell types were excluded without adequate justification (e.g. SOM+ to PV+). In addition, the construction of the afferent drive to the network does not reflect the stimulus presentations that are given in fear conditioning tasks. For instance, the authors only used a single training trial, the conditioning stimulus was tonic instead of pulsed, the unconditioned stimulus duration was artificially extended in time, and its delivery overlapped with the neutral stimulus, instead of following its offset. These deviations undercut the applicability of their findings.

This study partly achieves its aim of understanding how networks of biophysically distinctive interneurons interact to generate nested rhythms that coordinate the spiking of principal neurons. What still remains to demonstrate is that this promotes plasticity for training protocols that emulate what is used in studies of fear conditioning.

Setting aside the issues with the conditioning protocol, the study offers a model for the generation of multiple rhythms in the BLA that is ripe for experimental testing. The most promising avenue would be in vivo experiments testing the role of local VIP+ neurons in the generation of slow theta. That would go a long way to resolving whether BLA theta is locally generated or inherited from medial prefrontal cortex or ventral hippocampus afferents.

The broader importance of this work is that it illustrates that we must examine the function of neurons not just in terms of their behavioral correlates, but by their effects on the microcircuit they are embedded within. No one cell type is instrumental in producing fear learning in the BLA. Each contributes to the orchestration of network activity to produce plasticity. Moreover, this study reinforces a growing literature highlighting the crucial role of theta and gamma rhythms in BLA function.

Reviewer #2 (Public Review):

The authors of this study have investigated how oscillations may promote fear learning using a network model. They distinguished three types of rhythmic activities and implemented an STDP rule to the network aiming to understand the mechanisms underlying fear learning in the BLA. My comments are the following.

(1) Gamma oscillations are generated locally; thus, it is appropriate to model in any cortical structure. However, the generation of theta rhythms is based on the interplay of many brain areas therefore local circuits may not be sufficient to model these oscillations. Moreover, to generate the classical theta, a laminal structure arrangement is needed (where neurons form layers like in the hippocampus and cortex)(Buzsaki, 2002), which is clearly not present in the BLA. To date, I am not aware of any study which has demonstrated that theta is generated in the BLA. All studies that recorded theta in the BLA performed the recordings referenced to a ground electrode far away from the BLA, an approach that can easily pick up volume conducted theta rhythm generated e.g., in the hippocampus or other layered cortical structure. To clarify whether theta rhythm can be generated locally, one should have conducted recordings referenced to a local channel (see Lalla et al., 2017 eNeuro). In summary, at present, there is no evidence that theta can be generated locally within the BLA. Though, there can be BLA neurons, firing of which shows theta rhythmicity, e.g., driven by hippocampal afferents at theta rhythm, this does not mean that theta rhythm per se can be generated within the BLA as the structure of the BLA does not support generation of rhythmic current dipoles. This questions the rationale of using theta as a proxy for BLA network function which does not necessarily reflect the population activity of local principal neurons in contrast to that seen in the hippocampus.

(2) The authors distinguished low and high theta. This may be misleading, as the low theta they refer to is basically a respiratory-driven rhythm typically present during an attentive state (Karalis and Sirota, 2022; Bagur et al., 2021, etc.). Thus, it would be more appropriate to use breathing-driven oscillations instead of low theta. Again, this rhythm is not generated by the BLA circuits, but by volume conducted into this region. Yet, the firing of BLA neurons can still be entrained by this oscillation. I think it is important to emphasize the difference.

(3) The authors implemented three interneuron types in their model, ignoring a large fraction of GABAergic cells present in the BLA (Vereczki et al., 2021). Recently, the microcircuit organization of the BLA has been more thoroughly uncovered, including connectivity details for PV interneurons, firing features of neurochemically identified interneurons (instead of mRNA expression-based identification, Sosulina et al., 2010), synaptic properties between distinct interneuron types as well as principal cells and interneurons using paired recordings. These recent findings would be vital to incorporate into the model instead of using results obtained in the hippocampus and neocortex. I am not sure that a realistic model can be achieved by excluding many interneuron types.

(4) The authors set the reversal potential of GABA-A receptor-mediated currents to -80 mV. What was the rationale for choosing this value? The reversal potential of IPSCs has been found to be -54 mV in fast-spiking (i.e., parvalbumin) interneurons and around -72 mV in principal cells (Martina et al., 2001, Veres et al., 2017).

(5) Proposing neuropeptide VIP as a key factor for learning is interesting. Though, it is not clear why this peptide is more important in fear learning in comparison to SST and CCK, which are also abundant in the BLA and can effectively regulate the circuit operation in cortical areas.

eLife assessment

This useful modeling study explores how the biophysical properties of interneuron subtypes in the basolateral amygdala enable them to produce nested oscillations whose interactions facilitate functions such as spike-timing-dependent plasticity. The strength of evidence is currently viewed as incomplete because of insufficient grounding in prior experimental results and insufficient consideration of alternative explanations. This work will be of interest to investigators studying circuit mechanisms of fear conditioning as well as rhythms in the basolateral amygdala.

Reviewer #1 (Public Review):

Plasticity in the basolateral amygdala (BLA) is thought to underlie the formation of associative memories between neutral and aversive stimuli, i.e. fear memory. Concomitantly, fear learning modifies the expression of BLA theta rhythms, which may be supported by local interneurons. Several of these interneuron subtypes, PV+, SOM+, and VIP+, have been implicated in the acquisition of fear memory. However, it was unclear how they might act synergistically to produce BLA rhythms that structure the spiking of principal neurons so as to promote plasticity. Cattani et al. explored this question using small network models of biophysically detailed interneurons and principal neurons.

Using this approach, the authors had four principal findings:

(1) Intrinsic conductances in VIP+ interneurons generate a slow theta rhythm that periodically inhibits PV+ and SOM+ interneurons, while disinhibiting principal neurons.<br /> (2) A gamma rhythm arising from the interaction between PV+ and principal neurons establishes the precise timing needed for spike-timing-dependent plasticity.<br /> (3) Removal of any of the interneuron subtypes abolishes conditioning-related plasticity.<br /> (4) Learning-related changes in principal cell connectivity enhance expression of slow theta in the local field potential.

The strength of this work is that it explores the role of multiple interneuron subtypes in the formation of associative plasticity in the basolateral amygdala. The authors use biophysically detailed cell models that capture many of their core electrophysiological features, which helps translate their results into concrete hypotheses that can be tested in vivo. Moreover, they try to align the connectivity and afferent drive of their model with those found experimentally.

Deficient in this study is the construction of the afferent drive to the network, which does elicit activities that are consistent with those observed to similar stimuli. It still remains to be demonstrated that their mechanism promotes plasticity for training protocols that emulate the kinds of activities observed in the BLA during fear conditioning.

Setting aside the issues with the conditioning protocol, the study offers a model for the generation of multiple rhythms in the BLA that is ripe for experimental testing. The most promising avenue would be in vivo experiments testing the role of local VIP+ neurons in the generation of slow theta. That would go a long way to resolving whether BLA theta is locally generated or inherited from medial prefrontal cortex or ventral hippocampus afferents.

The broader importance of this work is that it illustrates that we must examine the function of neurons not just in terms of their behavioral correlates, but by their effects on the microcircuit they are embedded within. No one cell type is instrumental in producing fear learning in the BLA. Each contributes to the orchestration of network activity to produce plasticity. Moreover, this study reinforces a growing literature highlighting the crucial role of theta and gamma rhythms in BLA function.

Reviewer #2 (Public Review):

The authors of this study have investigated how oscillations may promote fear learning using a network model. They distinguished three types of rhythmic activities and implemented an STDP rule to the network aiming to understand the mechanisms underlying fear learning in the BLA.

After the revision, the fundamental question, namely, whether the BLA networks can or cannot intrinsically generate any theta rhythms, is still unanswered. The author added this sentence to the revised version: "A recent experimental paper, (Antonoudiou et al., 2022), suggests that the BLA can intrinsically generate theta oscillations (3-12 Hz) detectable by LFP recordings under certain conditions, such as reduced inhibitory tone." In the cited paper, the authors studied gamma oscillations, and when they applied 10 uM Gabazine to the BLA slices observed rhythmic oscillations at theta frequencies. 10 uM Gabazine does not reduce the GABA-A receptor-mediated inhibition but eliminates it, resulting in rhythmic populations burst driven solely by excitatory cells. Thus, the results by Antonoudiou et al., 2022 contrast with, and do not support, the present study, which claims that rhythmic oscillations in the BLA depend on the function of interneurons. Thus, there is still no convincing evidence that BLA circuits can intrinsically generate theta oscillations in intact brain or acute slices. If one extrapolates from the hippocampal studies, then this is not surprising, as the hippocampal theta depends on extra-hippocampal inputs, including, but not limited to the entorhinal afferents and medial septal projections (see Buzsaki, 2002). Similarly, respiratory related 4 Hz oscillations are also driven by extrinsic inputs. Therefore, at present, it is unclear which kind of physiologically relevant theta rhythm in the BLA networks has been modelled.

are there anyone with a value like this ?

Progression free survival seems to be missing ??

It would be useful to describe the causes and times of death for these patients as these are considered treatment related. 8.5% treatment related death is a significant proportion.

In the plot above please indicate that time is in years .

Dose intensity is not exactly the same as dose reduction. If the indicator asks for intensity then this should be clarified properly.

How was this variable entered ?

The difference of 1 patient for what reason ?

May be useful to report the maximum number of days. Also check the minimum number of days. Not clear as to what proportion have completed planned radiotherapy (may be 100%) - should be mentioned in the table.

Presumably these 3 patients did not start RT due to some reason here ?

Fornt line treatment duration is not calculated.

This is a very large gap and reasons for these need to be discussed including the protocol in place for starting chemotherapy.

Not clear from the table below what proportion of patietns did not have chemotherapy planned. Would be useful to have this percentage written though I presume that the 5 patient with chemo start date being unknown are the ones where chemotherapy was not planned.

For this data it would be quite useful to also include a histogram showing the distribution of days as some paitents have a very large number of days. Would be useful to discuss here what was the number of days post registration at TMC that RT was started. This would help clarify the issues related to delay in referrals.

laboratory

Check if any one had a negative date.

Needs harmonization.

This means all thre lineages had grade 3 toxicity. Were the anemia, leucopenia, lymphopenia, neutropenia and thrombocytopenia etc separated.

Does not make sense. I think it should < 3 year old ??

Similarly needs harmonization

Thsese seem to be same. This section need harmonization.

Write Eligible Instead

• Work toward a collection of use-cases
• Always design & test against all of them

• Introduction to Presentation

  • Speaker introduces themselves as Lou or Luke and begins the presentation titled "Cell Pond: Spatial Programming Without Escape."
  • The presentation is structured into three parts: spatial programming, Cell Pond, and the concept of programming without escape.

• Part 1: Spatial Programming

  • Lou shares a personal story from 23 years ago about their interest in trains and computers, learning to code with Stagecast Creator.
  • Stagecast Creator involved drag-and-drop spatial rules, where actions were defined by before-and-after pictures, such as character movement and obstacle navigation.
  • Despite its popularity, spatial programming had limitations, especially in 3D environments, requiring many rules for simple tasks.
  • Block-based coding became the norm, leading to a decline in spatial programming's popularity.

• Spatial Programming Tools

  • Various spatial programming tools are introduced, such as Splats, Color Code, NetBoden, and Viskit, each serving different purposes like education, art, simulation, and games.
  • Lou introduces their own tool, Sand Pond, which uses before-and-after pictures to code elements like sand, water, and rock.

• Challenges of Spatial Programming

  • The main challenge is the reliance on escape hatches, where more complex tasks require traditional programming, not just spatial rules.
  • Tools like Stagecast Creator and others incorporate lines of code for advanced functions, highlighting the limitation of pure spatial programming.

• Part 2: Cell Pond

  • Lou demonstrates Cell Pond, a tool where users can paint and create rules using color channels and directional triangles to manipulate spatial elements.
  • Example: A rule turning green into pink is shown, highlighting how spatial rules work dynamically.
  • Users can create complex simulations with multiple rules, illustrating conservation of properties like color across rules.
  • A key feature is creating generalized rules that apply to multiple elements, such as sand, water, and rock.

• Concept of Conservation in Cell Pond

  • Conservation ensures that properties remain consistent when elements interact, demonstrated through color copying.
  • Complex behaviors like diffusion are shown, where rules allow elements to swap positions randomly.

• Advanced Features of Cell Pond

  • Cell Pond supports splitting and merging cells to increase or decrease resolution and store more data.
  • Example: Creating a fractal by continuously splitting cells, demonstrating Cell Pond's potential for complex simulations.

• Part 3: Programming Without Escape

  • Lou questions if it's possible to create a powerful spatial programming tool without escape hatches.
  • The breakthrough came from creating a virtual machine called Dragon, which includes instructions for memory operations, streamlining the creation of spatial rules.

• Implications and Future Directions

  • Lou reflects on how the virtual machine (VM) simplifies the process, leading to better UI development.
  • The goal is to develop higher-level spatial programming languages, exploring what a "C of spatial programming" might look like.

• Q&A Session

  • Lou discusses the iterative design process and the importance of strict goals to achieve the intended functionality of Cell Pond.
  • Comparisons to other games like "Falling Sand" and "Baba is You" are made, highlighting influences and potential future directions.
  • Lou addresses the balance between adding UI features and maintaining VM capabilities, emphasizing the VM's core role in enabling advanced functions.
  • The presentation concludes with a discussion on live programming's role in creativity and how it fosters exploration and discovery in computational creativity.

line 6 supposed to print out 2 right? work 2/3 or do it in your head and its 2 lol not 3 (as remainder)

so int/int in this case 2/3 is gon be 0 cuz we see that in mathematics : 2/3 = 0.6 repeated but this is an int value cuz 2 and 3, therefore every number after decimal would be excluded and this process is the blue

line 7: score = score + 2;

line 7: score = score + 2;

台南空港

Before You Buy a Typewriter … Six Top Things You Need to Know by [[CreateX3]]

Designer Marcelo Nizzoli designed the Lettera 22, 32, and Studio 44.

Considerations for buying a typewriter:

1. Where will you use it? Is weight a consideration? (Standard, portable, ultra portable)
2. What Kind of Work will you do? (Do you need tabs, bichrome?)
3. How does the Typewriter Feel?
4. What type of keyboard will you need? (language, characters, diacritics)
5. What is the typeface? You should like/enjoy it and it should be big enough to be easily readable to you.
6. Do you like the typewriter's look/style?

How to test a typewriter before buying by [[Retrotype]]

hhh

https://danallosso.substack.com/p/hypothesis-social-and-private-annotation-053

Fun to see Dan Allosso using Hypothes.is as a more social media-related application instead of just the social annotation tool as many are using this in academia. It requires some additional work, but the discovery functionality is fantastic.

Voor en na de menopauze: De laatste menstruatie wordt ook wel menopauze genoemd. Voor en na de menopauze is er een periode van enkele jaren waarin de hormonen een nieuw evenwicht zoeken. Deze periode wordt de overgang, het climacterium, genoemd; de duur ervan is voor iedere vrouw verschillend.

Flebectomie = strippen

EVLT: endoveneuze lasertherapie: https://www.mcwetering.nl/spataderen/behandelingen/evlt-methode/

Sclerocompressie: https://www.mcwetering.nl/spataderen/behandelingen/sclerocompressie-therapie/

flebectomie https://www.mcwetering.nl/spataderen/behandelingen/flebectomie/?gad_source=1&gclid=CjwKCAjwhIS0BhBqEiwADAUhc0FmaH8CUCmQlTvoYqhbJ1cOgT4EM35YyYStZHp-nD6Iw3Pt6bdmshoCK94QAvD_BwE

Rule of Notice: First and last sentences are privileged, Rules of Signification: recognizing the meaning of the details we do notice. Knowing when to read a detail metaphorically, Rules of Configuration: help us to figure out what genre a text belongs to. Rules of Coherence: help us to figure out what genre a text belongs to.

When a shared resource is assigned to staffing profile, a single line is added to the top of staffing profile which represent the name of the shared resource and his or her allocations on this staffing profile. there is an indicator in front of the resource name to differentiate this shared resource from normal positions and assignments.

In the shared resource panel, the shared resources are grouped by Team. You can search for resources by name. how to: input the keyword and press Enter key.

when the shared resource is added to staffing profile, the remaining capacity of this resource is assigned to staffing profile by default. Project manager can save staffing profile directly. and they also can modify the assigned effort in Gantt or Table view before saving staffing profile.

Authenticating with the API

Now that we have set permissions on the API, it's essential to authenticate our requests if we want to perform actions like creating, updating, or deleting snippets.

Default Authentication Classes

By default, Django REST Framework uses the following authentication classes: - SessionAuthentication: Uses Django's session framework. - BasicAuthentication: Uses HTTP Basic Authentication.

When interacting with the API through a web browser, logging in through the browser session provides the required authentication for subsequent requests. However, for programmatic interaction, you need to explicitly include authentication credentials with each request.

Example of an Unauthenticated Request

If you try to create a snippet without authenticating, you will receive an error:

Unauthenticated Request Example

bash http POST http://127.0.0.1:8000/snippets/ code="print(123)"

Response

json { "detail": "Authentication credentials were not provided." }

Example of an Authenticated Request

To make an authenticated request, include the username and password of one of the users you created earlier.

Authenticated Request Example

Using httpie (a command-line HTTP client):

bash http -a admin:password123 POST http://127.0.0.1:8000/snippets/ code="print(789)" title="foo"

Response

json { "id": 1, "owner": "admin", "title": "foo", "code": "print(789)", "linenos": false, "language": "python", "style": "friendly" }

Summary

• Permissions: The API now has fine-grained permissions to ensure only authenticated users can create, update, or delete snippets.
• Authentication: Use either session-based or basic authentication for interacting with the API.
• Programmatic Access: Include the username and password in the request to authenticate programmatically.

Next Steps

In the next part of the tutorial, we'll: - Create an HTML endpoint for highlighted snippets. - Enhance the cohesion of the API by using hyperlinking for the relationships within the system.

This will further improve the usability and functionality of our web API, making it more intuitive and user-friendly.

To ensure that all code snippets are visible to everyone but only the user who created a snippet can update or delete it, you can create a custom permission class. This custom permission will be added to the snippet instance endpoint to enforce these rules.

Step-by-Step Instructions

1. Create Custom Permission Class: Create a new file called permissions.py in your snippets app directory and define a custom permission class IsOwnerOrReadOnly.

2. Update View to Use Custom Permission: Modify the SnippetDetail view to use the custom permission class in addition to the IsAuthenticatedOrReadOnly permission class.

Step 1: Create Custom Permission Class

snippets/permissions.py

```python from rest_framework import permissions

class IsOwnerOrReadOnly(permissions.BasePermission): """ Custom permission to only allow owners of an object to edit it. """

def has_object_permission(self, request, view, obj):
    # Read permissions are allowed to any request,
    # so we'll always allow GET, HEAD or OPTIONS requests.
    if request.method in permissions.SAFE_METHODS:
        return True

    # Write permissions are only allowed to the owner of the snippet.
    return obj.owner == request.user

```

Step 2: Update the View to Use Custom Permission

views.py

First, import the custom permission class at the top of your views.py file:

python from snippets.permissions import IsOwnerOrReadOnly

Then, update the SnippetDetail view to include the custom permission in the permission_classes property:

```python from rest_framework import generics from rest_framework import permissions from .models import Snippet from .serializers import SnippetSerializer

class SnippetList(generics.ListCreateAPIView): queryset = Snippet.objects.all() serializer_class = SnippetSerializer permission_classes = [permissions.IsAuthenticatedOrReadOnly]

def perform_create(self, serializer):
    serializer.save(owner=self.request.user)

class SnippetDetail(generics.RetrieveUpdateDestroyAPIView): queryset = Snippet.objects.all() serializer_class = SnippetSerializer permission_classes = [permissions.IsAuthenticatedOrReadOnly, IsOwnerOrReadOnly] ```

Explanation of the Code

• Custom Permission Class (IsOwnerOrReadOnly):
• permissions.BasePermission: This is the base class for all permissions in Django REST Framework.

• has_object_permission: This method checks whether the request has the required permissions for a specific object.

  • Read Permissions: Always allow safe methods (GET, HEAD, OPTIONS).
  • Write Permissions: Only allow if the user making the request is the owner of the object.

• SnippetDetail View:

• permission_classes: Combines IsAuthenticatedOrReadOnly (which allows read access to everyone and write access only to authenticated users) with IsOwnerOrReadOnly (which restricts write access to the owner of the snippet).

What Happens Now

• Read Access: Any user (authenticated or not) can read (list and retrieve) snippets.
• Write Access: Only authenticated users can create snippets, and only the owner of a snippet can update or delete it.

Testing the Setup

1. Open the Browsable API: Navigate to a snippet instance endpoint in your browser.
2. Check Actions: You should see the 'DELETE' and 'PUT' actions only if you are logged in as the user who created the snippet.

Summary

• Purpose: Ensure all code snippets are visible to everyone, but only the creator can update or delete their snippets.
• Implementation: Create a custom permission class and apply it to the SnippetDetail view.
• Result: Proper access control is enforced, allowing only the snippet owner to modify their snippet while everyone can read the snippets.

This setup ensures that your API adheres to the required permissions, providing both visibility and security.

To add login functionality to the browsable API in Django REST Framework, you need to include the authentication URLs in your project’s urls.py file. This will allow users to log in and log out via the browsable API interface.

Step-by-Step Instructions

1. Import the Required Modules: Add the necessary imports at the top of your urls.py file.
2. Include the Authentication URLs: Add a URL pattern to include the login and logout views for the browsable API.

Code Example

project-level urls.py

First, import the necessary modules:

python from django.urls import path, include

Then, add the authentication URL pattern:

```python from django.contrib import admin from django.urls import path, include

urlpatterns = [ path('admin/', admin.site.urls), path('api/', include('your_app.urls')), # Include your app's URLs path('api-auth/', include('rest_framework.urls')), # Add this line ] ```

Explanation of the Code

• path('api-auth/', include('rest_framework.urls')): This line adds the authentication URLs provided by Django REST Framework. It allows users to log in and log out through the browsable API.

What Happens Now

• Login Link: When you navigate to the browsable API in your browser, you will see a "Login" link in the top right corner.
• Login and Logout: Clicking on the "Login" link will take you to a login page where you can enter your credentials to log in. Once logged in, you can create, update, and delete snippets if you have the necessary permissions.

Testing the Setup

1. Open the Browsable API: Navigate to the browsable API in your browser.
2. Login: Click on the "Login" link in the top right corner and log in with a user account.
3. Create Snippets: Once logged in, you will be able to create new code snippets and perform other actions that require authentication.

Example

Let's assume you have already created a few users and code snippets. After logging in as one of these users, you can create a new snippet via the browsable API.

Creating a New Snippet

1. Navigate to the endpoint for creating snippets (e.g., /api/snippets/).
2. Fill in the details for the new snippet and submit the form.

Viewing Users

Navigate to the /api/users/ endpoint. The representation will include a list of snippet IDs associated with each user in the snippets field.

Summary

• Purpose: Adding login functionality to the browsable API allows users to authenticate and perform actions that require login.
• Implementation: Include the rest_framework.urls in your urls.py file.
• Result: Users can log in and log out via the browsable API, enabling them to create, update, and delete snippets.

This setup enhances the usability of your API, making it easier for users to interact with it directly through the browser.

To ensure that only authenticated users can create, update, or delete code snippets while allowing unauthenticated users to read the snippets, we can use the IsAuthenticatedOrReadOnly permission class from Django REST Framework.

Here's how you can implement this in your views:

Step-by-Step Instructions

1. Import Permissions: First, import the permissions module from Django REST Framework.
2. Set Permission Classes: Add the permission_classes property to both the SnippetList and SnippetDetail view classes, setting it to IsAuthenticatedOrReadOnly.

Code Example

views.py

First, import the permissions at the top of your views.py file:

python from rest_framework import permissions

Then, update your view classes to include the permission_classes property:

```python from rest_framework import generics from .models import Snippet from .serializers import SnippetSerializer

class SnippetList(generics.ListCreateAPIView): queryset = Snippet.objects.all() serializer_class = SnippetSerializer permission_classes = [permissions.IsAuthenticatedOrReadOnly]

def perform_create(self, serializer):
    serializer.save(owner=self.request.user)

class SnippetDetail(generics.RetrieveUpdateDestroyAPIView): queryset = Snippet.objects.all() serializer_class = SnippetSerializer permission_classes = [permissions.IsAuthenticatedOrReadOnly] ```

Explanation of the Code

• permission_classes: This property specifies the permissions that are required to access the view.
• permissions.IsAuthenticatedOrReadOnly: This permission class ensures that:
• Authenticated users (logged in) can perform any action (read, create, update, delete).
• Unauthenticated users (not logged in) can only read the data (list and retrieve).

What Happens Now

• Authenticated Users:
• Can create new snippets.
• Can update existing snippets.
• Can delete snippets.
• Can read (list and retrieve) snippets.
• Unauthenticated Users:
• Can only read (list and retrieve) snippets.
• Cannot create new snippets.
• Cannot update snippets.
• Cannot delete snippets.

Example Usage

As an Authenticated User

When a logged-in user sends a POST request to create a snippet, it will be allowed because they have read-write access.

bash curl -X POST -H "Authorization: Token <user-token>" -d '{"title": "New Snippet", "code": "print(123)"}' http://example.com/snippets/

As an Unauthenticated User

When a not logged-in user tries to send a POST request to create a snippet, it will be denied because they only have read-only access.

bash curl -X POST -d '{"title": "New Snippet", "code": "print(123)"}' http://example.com/snippets/

This request will result in a 403 Forbidden response, indicating that the user does not have permission to perform the action.

Summary

• Purpose: To restrict create, update, and delete actions to authenticated users, while allowing unauthenticated users to read data.
• Implementation: Use the IsAuthenticatedOrReadOnly permission class in the view classes.
• Effect: Authenticated users get full access, while unauthenticated users get read-only access.

This setup helps protect your data by ensuring that only users who are logged in can modify it, while still allowing anyone to view the data.

Let's break down how to update the SnippetSerializer to include the owner field and explain what this change does in simple terms.

Step-by-Step Explanation

1. Add the owner Field: In your SnippetSerializer, add a new field called owner that will show the username of the user who created the snippet.

2. Update the Meta Class: Make sure to include the owner field in the list of fields in the serializer's Meta class.

Code Example

Here's how you can update your SnippetSerializer:

```python from rest_framework import serializers from .models import Snippet

class SnippetSerializer(serializers.ModelSerializer): owner = serializers.ReadOnlyField(source='owner.username')

class Meta:
    model = Snippet
    fields = ['id', 'title', 'code', 'linenos', 'language', 'style', 'owner']

```

Explanation of the Code

• owner Field:
• serializers.ReadOnlyField: This type of field is read-only, meaning it is only used when the data is being sent out, not when data is being received.
• source='owner.username': The source argument specifies which attribute to use to fill this field. owner.username means it will use the username attribute of the owner (the user who created the snippet).

What This Field Does

• Read-Only: The owner field is read-only, so it will show up when you serialize a snippet, but it won't be used when you create or update a snippet.
• Source Attribute: The source attribute lets you specify which attribute of the model to use. In this case, owner.username will use the username of the user who owns the snippet.

How It Works

When you serialize a Snippet instance, the owner field will include the username of the user who created it.

Example

Let's say we have a user named Alice who has created a snippet. When we serialize that snippet, it will look like this:

```python

Assuming snippet is an instance of Snippet created by user 'Alice'

serializer = SnippetSerializer(snippet) print(serializer.data) ```

Output

json { "id": 1, "title": "Example Snippet", "code": "print('Hello, world!')", "linenos": true, "language": "python", "style": "friendly", "owner": "Alice" }

Summary

• Purpose: Adding the owner field to the SnippetSerializer allows us to show the username of the user who created each snippet.
• Read-Only: The owner field is read-only, meaning it's used for displaying data but not for creating or updating snippets.
• Source Attribute: The source='owner.username' part ensures that the field will display the username of the owner.

This update makes your API responses more informative by including the creator's username with each snippet, providing more context and making it easier to understand who created each snippet.

Cette vidéo présente l’approche de l’enseignement explicite, qui est une pédagogie basée sur les recherches sur l’enseignement efficace. Elle explique les principes, les étapes et les avantages de cette méthode, ainsi que les critères pour choisir le niveau de guidage approprié selon le contexte.

Résumé vidéo [00:00:03][^1^][1] - [00:24:49][^2^][2] :

Cette vidéo présente l'approche de l'enseignement explicite, qui est une pédagogie basée sur les recherches sur l'enseignement efficace. Elle explique les principes, les étapes et les avantages de cette méthode, ainsi que les critères pour choisir le niveau de guidage approprié selon le contexte.

Points forts : + [00:00:03][^3^][3] L'origine de l'enseignement explicite * Basé sur l'observation des enseignants qui faisaient le plus progresser les élèves * Développé par Barak Rosenshine à partir des stratégies identifiées comme efficaces * Vise à rendre visible la compréhension et à éviter l'implicite + [00:06:53][^4^][4] Les étapes de l'enseignement explicite * Le modelage : l'enseignant montre et explique le raisonnement, les stratégies, les liens, etc. * La pratique guidée : l'enseignant vérifie la compréhension des élèves, les fait pratiquer en binôme ou en groupe, leur fournit du feedback et de l'étayage * La pratique autonome : l'enseignant laisse les élèves faire les tâches seuls, avec un taux de succès élevé + [00:18:10][^5^][5] Les critères pour choisir le niveau de guidage * Le niveau de compétence des élèves : plus il y a d'élèves en difficulté, plus le guidage doit être soutenu * Le niveau de nouveauté et de complexité de la tâche : plus la tâche est nouvelle et complexe, plus le guidage doit être explicite * Le temps dont on dispose : plus on utilise des approches par découverte, plus le temps sera long * Le type de contenu : les idées maîtresses du programme doivent être enseignées de façon explicite, les idées secondaires peuvent être plus découvertes

Résumé vidéo [00:24:51][^1^][1] - [00:49:14][^2^][2]:

Cette partie de la vidéo parle de l'enseignement explicite et du modèle de réponse à l'intervention (RI) pour aider les élèves en difficulté d'apprentissage ou de comportement. Il explique les origines, les principes et les exemples de ces approches pédagogiques fondées sur des données probantes.

Points forts: + [00:24:51][^3^][3] L'enseignement explicite * Une pédagogie adaptée aux besoins des élèves * Une guidance explicite, structurée et systématique * Un modelage, une pratique guidée, une pratique autonome et une consolidation + [00:26:00][^4^][4] La réponse à l'intervention (RI) * Un modèle inspiré de la santé publique * Trois paliers d'intervention selon le degré de difficulté des élèves * Des pratiques efficaces et une mesure constante des progrès + [00:38:05][^5^][5] Des exemples concrets de RI * La gestion des comportements dans les corridors, les classes, la cafétéria, etc. * L'enseignement de la lecture, de l'écriture, des mathématiques, etc. * La collaboration entre les enseignants, les spécialistes et les chercheurs

there

Résumé de la vidéo [00:00:00][^1^][1] - [00:24:48][^2^][2]:

Cette vidéo explore pourquoi certains réussissent en mathématiques plus facilement que d'autres. Elle aborde les mythes et les réalités de l'enseignement des mathématiques en France, en mettant l'accent sur les méthodes pédagogiques et les inégalités.

Moments forts : + [00:00:00][^3^][3] Démystification des idées reçues * Les Français ne sont pas tous mauvais en maths * Les filles peuvent être aussi bonnes en maths que les garçons * La réussite en maths n'assure pas la réussite scolaire + [00:01:32][^4^][4] Importance des mathématiques dans l'éducation * Les mathématiques occupent une place centrale dans les politiques éducatives * Le plan de 21 mesures pour l'enseignement des mathématiques en France * Les résultats PISA et leur interprétation + [00:03:00][^5^][5] Analyse des résultats PISA * La France est dans la moyenne de l'OCDE * Différences de résultats entre élèves favorisés et défavorisés * Les filles sont plus anxieuses que les garçons en France + [00:06:00][^6^][6] Inégalités de genre en mathématiques * Différences de résultats dès la fin du CP * Importance de présenter des modèles féminins réussis en mathématiques * Impact des stéréotypes sur la perception des compétences en maths + [00:09:00][^7^][7] Méthode Singapour et autres approches pédagogiques * Débat sur l'adoption de la méthode Singapour en France * Importance de la formation des enseignants * Verbalisation et manipulation comme clés de l'apprentissage

Est-ce que tu as trouvé ces points intéressants ?

Résumé de la vidéo [00:24:51][^1^][1] - [00:47:48][^2^][2]:

Cette vidéo explore pourquoi certains réussissent en mathématiques plus facilement que d'autres, en abordant des sujets tels que les inégalités sociales, l'anxiété liée aux mathématiques, et l'importance de la méthode d'enseignement.

Moments forts: + [00:24:51][^3^][3] Inégalités en mathématiques * Progrès à faire sur les inégalités sociales * Importance des modèles positifs pour les filles * Anxiété liée aux mathématiques + [00:26:00][^4^][4] Sélection par les mathématiques * Pression sociale et scolaire * Sélection par les mathématiques pour certaines carrières * Critique de la sélection par les mathématiques + [00:30:02][^5^][5] Neurosciences et apprentissage * Importance des séances courtes et régulières * Courbe de l'oubli et rappel épisodique * Variables d'apprentissage adaptées + [00:35:01][^6^][6] Évaluation et répétition * Évaluations courtes et régulières * Importance de l'entraînement dosé * Éviter les évaluations stressantes + [00:40:01][^7^][7] Formation des enseignants * Importance de la culture mathématique * Formation des enseignants en mathématiques * Rassurer les enseignants sur l'enseignement des mathématiques

Est-ce que cela répond à votre demande?