The conserved biochemical activity and function of an early metazoan phosphatidylinositol 5 phosphate 4-kinase regulates growth and development

  1. Harini Krishnan
  2. Suhail Muzaffar
  3. Sanjeev Sharma
  4. Visvanathan Ramya
  5. Avishek Ghosh
  6. Ramanathan Sowdhamini
  7. Padinjat Raghu

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

The ability to co-ordinate function between multiple cells is a critical requirement for multi-cellularity. This co-ordination is mediated by hormones or growth factors, molecules secreted by one cell type that can convey information to the other cells and influence their behaviour. Hormone-dependent signalling is mediated by second messenger systems;phosphoinositides (PIs) generated by lipid kinase activity are one such key second messenger system. Phosphatidylinositol 5 phosphate 4-kinase (PIP4K) is a lipid kinase that phosphorylates phosphatidylinositol 5-phosphate (PI5P) to generate phosphatidylinositol 4,5 bisphosphate [PI(4,5)P 2 ]. Following a comprehensive bioinformatics analysis of ca. 23296 proteomes covering the tree of life, we find that PIP4K is a metazoan-specific enzyme, although its homologs are also found in choanoflagellate genomes. To understand their function in early metazoans, we experimentally analysed the biochemical activity and physiological function of PIP4K from several early metazoans. We find that the PIP4K enzyme from an early branching metazoan sponge Amphimedon queenslandica (AqPIP4K), regarded as the earliest evolved metazoan, shows a biochemical activity highly conserved with human PIP4K; AqPIP4K is able to selectively phosphorylate PI5P to generate PI(4,5)P 2 just as effectively as the human enzyme. Further, AqPIP4K was able to rescue the reduced cell size, growth and development phenotype in larvae of a null mutant in Drosophila PIP4K. These phenotypes are regulated through activity of the hormone insulin, acting via the cell surface insulin receptor, a member of the receptor tyrosine kinase family, that is unique to metazoans. Together, our findings indicate that in early metazoans, AqPIP4K is likely to function in a signal transduction pathway that is required for receptor tyrosine kinase signalling. Overall, our work defines PIP4K as a signal transduction motif required to regulate receptor tyrosine kinase signalling for intercellular communication in the earliest forms of metazoa.

Article activity feed

  1. Review Commons

    Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    1. General Statements [optional]

    We thank the reviewers for their insightful comments regarding our study and for appreciating the range of experiments used, the depth of our study and the significance of our work. We also thank reviewers with expertise in evolutionary biology for highlighting the need for precise wrong of some parts of the manuscript and the need for balancing the various viewpoints on the current understanding of early metazoan evolution. A point-by-point response to each reviewer comment is given below. We believe that we can effectively address most reviewer comments in a revised version. The revised improved manuscript will be the first insightful study of intracellular signalling pathways in the context of early animal evolution. We thank the reviewer for noting that this study is highly impactful and can have a broader influence on the scientific community.

    2. Description of the planned revisions

    Reviewer #1 (Evidence, reproducibility and clarity (Required)):

    __ Summary: The researchers identified PIP4K (phosphatidylinositol 5 phosphate 4-kinase) as a lipid kinase that is specific to metazoans. In order to determine its conserved function across metazoans, they compared PIP4K activity in both early-branching metazoans and bilaterian animals. Biochemical assays demonstrated a conserved catalytic activity between the sponge Amphimedon queenslandica (AqPIP4K) and human PIP4K. In in-vivo experiments, AqPIP4K was found to rescue the reduced cell size, growth, and development phenotype in larvae of null mutant in Drosophila PIP4K. Based on these findings, the authors suggest that the function of PIP4K was established in early metazoans to facilitate intercellular communication. The experiments were well designed, and a range of biochemical, in vitro, and in vivo experiments were conducted.__

    __ That being said, there are some questions that require further discussion before we can fully accept the author's conclusion of an evolutionarily conserved function of PIP4K across metazoans.__

    Major comments:

    The authors mentioned that PIP4K is metazoan-specific and involved in intercellular communication. How can we explain the presence of PIP4K in choanoflagellate genomes? Despite its high similarity with conserved domains and functionally important residues, experimental results with the PIP4K from Choanoflagellate (Monosiga brevicollis, MbPIP4K) such as Mass spectrometry-based kinase assay and mutant Drosophila PIP4K didn't show similar activity to sponge AqPIP4K. The authors suggested that "In the context of other ancient PIP4K it is possible that since choanoflagellates exist as both single-cell and a transient multicellular state and do not have the characteristics of metazoans, PIP4K does not play any important functional role in these." However, this explanation is not well justified; they need to provide a more detailed discussion on this. Response: PIP4K is found in the genome of the choanoflagellate, M.brevicollis. MbPIP4K has the requisite kinase domain and the critical residue in the activation loop (A381) required for PIP4K activity is also conserved with the Amphimedon enzyme. Despite this, MbPIP4K was unable to rescue the growth and cell size phenotype of dPIP4K mutants (dPIP4K29) unlike AqPIP4K.

    We have previously published a comparison of the in vitro activity versus in vivo function for the three PIP4K enzymes in the human genome (Mathre et.al PMID: __30718367). While all three human PIP4K isoforms can functionally rescue the *Drosophila *dPIP4K mutant, there is a nearly 104-fold difference for in vitro activity between them with PIP4K2C showing almost no in vitro activity. __The difference in in vitro enzyme activity between MbPIP4K and AqPIP4K is similarly notable. We would however highlight that this is more likely a reflection of the limitations of the *in vitro *PIP4K activity.

    However, while AqPIP4K can rescue function in vivo (rescue of fly mutant phenotypes) MbPIP4K could not when expressed in fly cells. This must imply that there are differences in the polypeptide sequences of AqPIP4K and MbPIP4K that allow the former but not the latter to couple to the Insulin PI3K signalling pathway in fly cells. Given that Amphimedon and Choanoflagellates are separated by 100-150 Mya in evolution, this is possible. Our data on expression of AqPIP4K and MbPIP4K in fly S2 cells shows that they do not have equivalent localization (Fig 2C). What are the differences in the two polypeptides that lead to this? We will perform a multiple sequence alignment using PIP4K sequences from multiple choanoflagellates and sponges to identify these differences.

    We will include the results of this analysis and an appropriate discussion in the revised manuscript.

    • Likewise, the PIP4K gene has been identified in cnidarians, which are a sister group to bilaterian animals. However, the Cnidaria HvPIP4K showed no activity in biochemical or functional assays. In comparison to sponges, cnidarians are relatively complex organisms, and I believe that PIP4K is highly important for intercellular communication, as it is in bilaterians. The authors attempted to explain this by suggesting that "Based on theories of parallel evolution between cnidarians and sponges during early metazoan evolution, it is possible that the PIP4K gene was retained functional in one lineage and not in other." However, I am not convinced by this statement.

    Response: This is a really interesting and challenging question from the reviewer. We are aware that both sponges (Porifera) and Cnidaria are examples of primitive metazoans separated by 80-90 Mya of evolution, yet while AqPIP4K shows activity and can functionally rescue dPIP4K mutants, HvPIP4K cannot. What does this mean?

    A key difference between sponges and cnidarians is that while cnidarians have a simple “nerve-net” like nervous system, sponges do not have such a mode of communication. Therefore, it is possible that PIP4K, which we propose works in the context of hormone-based communication, is functionally important in sponges.

    We are of course aware and acknowledge that in a like for like experimental system (Drosophila cells) our data shows that the two proteins behave differently, be it in terms of in vitro activity or in vivo function. This must imply inherent differences in the two polypeptides.

    What we propose to do is to compare available PIP4K sequences from multiple Porifera and Cnidaria genomes and try and understand differences in the protein sequence that might explain differences in function. These results and their implications will be included in the revised manuscript.

    Please provide details of the databases (Uniprot-KB, NCBI sequence database, Pfam) versions. After identifying the specific PIP4K protein in each species (e.g. AqPIP4K and HvPIP4K), have you considered performing a reciprocal blast against the human genome to see if you have a top hit to PIP4K? Hence, the main focus of the project is on PIP4K as a metazoan-specific protein. We need to include a wider representation of non-bilaterian animals, including multiple species from sponges, ctenophores, placozoans, and cnidarians. Additionally, please check if homologues of PIP4K are present in other unicellular holozoans besides choanoflagellates. Response: We will add the NCBI IDs for all the sequences. We have carried out reciprocal blast to human proteome and then classified the selected sequences as PIP4K, we will add the results in the supplementary for the same. We will add more species of sponges, ctenophores, placozoans, and cnidarians in our analysis of PIP4K sequences. We will also include an analysis of other unicellular holozoans where genome sequence is available.

    Authors suggested the identification of other components of the PI signaling pathway along with PIP4k in the sponge. What is the status of these PI signaling pathway genes in other non-bilaterians and choanoflagellates? Response: We will add the details of the same in the revised manuscript and agree that this will help enhance the interpretation of our results.

    Phylogenetic tree of all PIP4K sequences (Figure 1C): How authors can be certain that the identified PIP4K sequences (e.g. AqPIP4K, HvPIP4K, and MbPIP4K) are indeed PIP4K, especially when there are several closely related proteins? It is important to conduct phylogenetic analysis alongside other PIP sequences (such as PI3K, PI4K, PIP5K, and PIP4K). If this analysis is carried out, the identified AqPIP4K, HvPIP4K, and MbPIP4K should be grouped together with human PIP4K in the same cluster. Response: As described in the methods, we have searched all the individual genomes analyzed for all PIK and PIPK enzyme sequences. We have marked the domains (using Pfam and Interpro) on these sequences and eliminated other PIK and PIPK sequences (such as PI3K, PI4K, PIP5K) and selected only PIP4K. To additionally confirm the distinction between PIP5K and PIP4K, we have manually inspected each sequence to establish the identity of the A381 amino acid residue in the activation loop. The identity of the amino acid at this position in the activation loop has been experimentally demonstrated to be an essential feature of PIP4K (Kunze et.al PMID: 11733501) and we have also confirmed this independently in a recent study (Ghosh et.al PMID: 37316298).

    We will perform the phylogenetic analysis of the phosphoinositide kinases in the format suggested by the reviewer and add it in the revision as a supporting evidence.

    Minor comments:

    Line 157: Phylogenetic conservation of PIP4Ks: Please provide details about bootstrap analysis. Response: Will be added

    Line 230: symbol correction 30{degree sign}C Response: Will be done

    Line 429-430: "from early metazoans like Sponges, Cnidaria and Nematodes." Nematodes are not considered early metazoans. Response: Apologies for the typo. This will be corrected. We agree that nematodes are not early metazoans.

    Line 477-478: "However, interestingly, MbPIP4K::GFP localizes only at the plasma membrane in S2 cells (Figure 2C)." This part was not further discussed. Can you please elaborate on why MbPIP4K::GFP localizes only at the plasma membrane in S2 cells? Response: We have discussed this point specifically in response to major comment by the reviewer and it will be addressed as described.

    Line 598: "the earliest examples of metazoa, namely the coral A. queenslandica" A. queenslandica is a sponge, not coral. Response: Apologies for the error. We will correct it.

    Line 602: "Amphimedon and human enzyme, although separated by 50Mya years of evolution" I think it's 500 million years ago, not 50 million years ago. Response: This typo will be corrected.

    Line 612: "coordinated communication between the cells is the most likely function" the cell. Response: Will change the sentence accordingly

    • Line 614: "intracellular phosphoinositide signalling the identity of the hormone" missing full stop punctuation. Response: Will change the sentence accordingly

    Line 802 - 804: "other by way of difference in colour. The sub clusters have been numbered (1- early metazoans, 2- Nematodes, 3- Arthropods, 4- Molluscs, 5- Vertebrates (isoform PIP4K2C), 6- Vertebrates (isoform PIP4K2A), 7- Vertebrates (isoform PIP4K2B)." In the Figure, I can't find numbers on the subclusters. Response: Will add the numbers in the figure.

    Line 805- 807: "Phylogenetic analysis of selected PIP4K sequences from model organisms of interest. PIP4K from A. queenslandica has been marked in rectangular box." The rectangular box is missing in the figure. Response: Will change the figure accordingly

    Figure 1C: full forms of species names are missing. Response: Will change the figure accordingly

    Reviewer #1 (Significance (Required):

    The data is presented well, and the authors used a wide range of assays to support their conclusion. The study is highly impactful and can have a broader influence on the scientific community, particularly in evolutionary molecular biology, development, and biochemistry.

    The study provides interesting findings; however, the reasons for PIP4K not being functional in cnidarians as in sponges and why PIP4K is present in unicellular holozoans but not functional are unclear.

    We thank the reviewer for appreciating the significance and impact of our study. The very helpful questions raised by the reviewer will help enhance the quality of our study even further. We will make every effort to address these queries.

    Reviewer #2 (Evidence, reproducibility and clarity (Required):

    The manuscript by Krishnan et al. uses molecular phylogenetics, in vitro kinase assays, heterologous expression assays in Drosophila S2 cells and mutant complementation assays in yeast to study the evolution and function of putative PIP4 kinase genes from a sponge, a cnidarian and a choanoflagellate. Based on these experiments, the authors conclude that PIP4K is metazoan-specific and that the sponge PIP4K has conserved functions in selectively phosphorylating PI5P.

    The study is in principle of interest and it could all be valid data, but the large number of flaws in the data presentation and/or analysis just makes it hard to assess the quality and thus validity of the data and conclusions.

    We thank the reviewer for appreciating the potential interest in our findings of PIP4K function in early metazoans. We thank them for noting the need for correcting data presentation and these will be done in the revision.

    __ Major comments:__

    Overall, the manuscript lacks scientific rigor in the analysis and representation of the results, and the validity of many of the conclusions is therefore difficult to assess.

    Major problems are:

    (i) The authors base their study on the evolution of PIP4K genes on a deeply flawed concept of animal evolution. On multiple occasions, including the title, the authors refer to extant species (e.g. Amphimedon) as 'early metazoan', 'regarded as the earliest evolved metazoan' (l. 46-7) or 'the earliest examples of metazoans' just to name a few. This reflects a 'ladder-like' view on evolution that suggests that extant sponges are identical to early 'steps' of animal evolution.

    We thank the reviewer who is clearly vastly more experienced in the field of evolutionary biology for the possible imprecise/incorrect usage of the word “ancient metazoan”. As new entrants to this area of evolutionary biology, we have of course referred to the existing literature such as PMID: 20686567 to guide us. This paper describes the sequencing of the *A. queenslandica *genome. It is clear that there is perceived value in studying this sponge in the context of early animal evolution although we are aware of there are a multitude of sponges and not all of them may be of value in the study of early animal evolution. We will peruse the literature more carefully and revise the manuscript to provide a more balanced view of this very interesting but unresolved area.

    Also, the author's interpretation that one cluster of genes 'contained the sequences from early metazoans like sponges, cnidaria and nematodes' is referring to an outdated idea of animal phylogeny where nematodes were thought to be ancestrally simple organisms grouped as 'Acoelomata'. This idea of animal phylogeny was however disproven by molecular phylogenetics since the 1990ies.

    Response: We are aware that the field of animal classification is undergoing continuous evolution. While earlier classifications may have been based of the presence or otherwise of a coelom and/or other anatomical features, we are aware of the use of molecular phylogenetics.

    The phylogeny presented in Fig 1C is based on the sequence relationships between the PIP4K sequences from various animal genomes. Any errors in the labelling of groups such as that highlighted by the reviewer will be revised or corrected after a careful consideration of extant views in the field, which are somewhat varied.

    (ii) The description of taxa in the phylogenetic tree in Fig. 1B lacks any understanding of phylogenetic relationships between animals and other eukaryotic groups. What kind of taxa are 'invertebrates' or 'parasites'? And why would 'invertebrates' exclude cnidarians and sponges? Also, why is the outgroup of opisthokonts named 'Eukaryota'?? Are not all organisms represented on the tree eukaryotes?

    Response: We apologize for this imprecision in labelling taxa. This will be corrected.

    (iii) The methods part lacks any information about the type of analysis (ML, Bayesian, Parsimony?) used to perform the phylogenetic analysis shown in Fig. 1C. Also, the authors mention three distinct clusters (l.428) that are not labelled in the figure.

    Response: We will update the methods to include the additional details requested by the reviewer. Fig 1C will be re-labelled.

    (iv) The validity of the Western Blot is difficult to assess as the authors have cut away the MW markers. Without, it is for example difficult to assess the size differences visible between Hydra and Monosiga PIP4K-GFP proteins on Fig. 2B. Also, it has become standard practice to show the whole Western blot as supplementary data in order to assess the correct size of the bands and the specificity of the antibody. This is also missing from this manuscript.

    Response: Cropped Western blots have been shown to facilitate figure preparation in the main manuscript. The complete uncropped Western blots, in all cases, will be shared as Source data as is the standard practice for multiple journals in the review Commons portfolio.

    (v) The authors claim that AqPIP4K was able to convert PI3P into PI with very low efficiency (Figure 2E), but without further label in the figure or explanation, it remains unclear how the authors come to this conclusion.

    Response: We regret the typo in line 500 of the manuscript we have stated that “Further,……… was able to convert PI3P into PI with very low efficiency (Figure 2E).” What we intended to write was “Further,……… was able to convert PI3P into PI (3,4) P2 with very low efficiency (Figure 2E).” The efficiency with which this reaction takes place is very low and has been reported by us (Ghosh et.al PMID: 31652444) and others (Zhang et.al PMID: 9211928). At the exposure of the TLC shown in Fig 2E the PI(3,4)P2 spot cannot been seen. Much longer exposures of the TLC plate will be needed to see the PI(3,4)P2 spot. This will be corrected in a revised version of the manuscript.

    (vi) The box plots in Fig. 3C and D lack error bars and thus seem to be consisting of only single data points without replicates. Also, Fig. 3C is a quantification of Fig. 3B but it remains unclear what has been quantified and how. It is also unclear how %PIP2 was determined.

    Response: For Fig 3C, the colony count has been done from three replicates and the average has been considered to calculate the % growth for each genotype. We will include error bars and clarify this in the revised figure legend. For Fig 3D, the PIP2/PIP ratio has been calculated from biological replicates and average has been represented in the graphs. The individual values can be provided as supplementary data.

    (vii) Throughout Fig. 4, I do not understand the genotypes indicated on the x-axis of the plots and below the images. I read the figure legends and manuscript describing these results at least 3 times, but cannot figure out what it all means. On Fig. 4C, what is the wild-type situation?

    Response: We apologize for the lack of precision in labelling the figures versus the figure legends. This will be corrected in the revision:

    The genotypes are as follows

    • *w1118 *(control) * Act-GAL4. This has been referred to as wild type in the figure legend and called Act-Gal4 in Fig4 panels A-E
    • dPIP4K29 – This refers to the protein null strain of dPIP4K. This strain is the background in which all reconstitutions of PIP4K genes have been done.
    • PIP4K transgene from A. queenslandica.
    • AqPIP4KKD Kinase dead PIP4K transgene from * queenslandica*. In panels A, B, D and E, Act-GAL4: dPIP4K29 indicates the genetic background in which either AqPIP4K or AqPIP4KKD has been reconstituted.

    Reviewer #2 (Significance (Required)):

    If validated and put in the right phylogenetic context, the study is potentially contributing to expanding our knowledge on the evolution of metazoan-specific features, especially the evolution of proteins involved in cell-cell signalling and growth control. My field of expertise is broadly in evo-devo, molecular phylogentics, developmental genetics and cell biology. The in vitro lipid analysis seems interesting and potentially valid but I do not have sufficient expertise to evaluate its validity.

    We thank the reviewer for appreciating the novelty of our contribution and its potential to contribute to understanding the evolution of metazoan specific signalling systems, once appropriate corrections have been made. We also appreciate their positive comment on our in vitro experimental analysis. This paper is a big effort to not only perform phylogenetic analysis but address the emerging interpretations experimentally as much as possible.

    Reviewer #3 (Evidence, reproducibility and clarity (Required)):

    Summary In this manuscript, the authors investigate the evolutionary origins of metazoan Phosphatidylinositol phosphates (PIPs) signaling by elucidating the sequence and function of the PIP4K enzyme, which is crucial for converting PI5P to PI(4,5)P2 through phosphorylation. The authors have described PIP4K-like sequences distributed throughout metazoans and choanoflagellates through an extensive sequence screening. With in vitro and in vivo functional assays, the authors have shown that the sponge A. queenslandica PIP4K (AqPIP4K) is functionally similar to its human counterpart and highlight the major discovery of this study - that PIP4K protein function dates back to as early as sponges.

    We thank the reviewer for noting the major finding of our study and our efforts to experimentally validate, using multiple approaches, the findings of our detailed bioinformatics analysis of PIP4K gene distribution across the tree of life.

    Major comments

    There are two key limitations to this paper. Like the sponges, ctenophores are one of the earliest branching metazoans. They are not well addressed in the paper. Secondly, despite finding PIP4 homologs in choanoflagellates, the authors claim that PIP4 is metazoan-specific.

    We thank the reviewer for highlighting these two points; we recognize that both of these are important to address, to the extent that it is possible to do so. These will be addressed using the approaches detailed in the response to reviewer 1 comments.

    Line 46: A. queenslandica is the earliest branching metazoan. The phylogeny of sponges and ctenophores is not conclusively defined and hence, the statement must be rephrased. Despite the brief description of the evolution of metazoan lineage in the discussion section, ctenophores are missing from the phylogenetic tree. At least a sequence-level information PIP4K in ctenophores would strongly back the claims of the manuscript. Here is the link to the Mnemiopsis database. Response: We thank the reviewer for highlighting this point and pointing us to the Mnemiopsis database. We will most certainly analyse ctenophore genome sequences and add the ctenophore PIP4K sequence to the phylogeny, post analysis and the discussion will be modified to reflect the findings.

    Mentioning that choanoflagellates contain homologs of PIP4K contradicts the statement that PIP4K is metazoan-specific. As per Fig 1E., the domain organization of PIP4K is conserved among choanoflagellates and metazoans. What is the percent sequence similarity to the query? This could answer why it doesn't show activity in Drosophila rescues - the system might simply not be compatible with the choanoflagellate homolog. The same may apply to the cnidarian homolog HvPIP4K. Further evidence is needed before concluding that MbPIP4K doesn't phosphorylate PIP5. It is additionally fascinating that MbPIP4K localizes at the plasma membrane unlike other homologs - this function might be choano-specific. Overall, PIP4K's possible origin in the choanoflagellate-metazoan common ancestor backs the current research that choanoflagellates indeed hold clues to understanding metazoan evolution. Further research is necessary before concluding (as in line 648) in the discussions section, where it is mentioned that "PIP4K does not play any important functional role in choanos".

    Response: We thank the reviewer for highlighting the very interesting but incompletely understood facets of our study vis-à-vis choanoflagellates versus metazoans. The proposal for additional analysis is indeed interesting and we will carry out these analysis and revise the text accordingly.

    __ Minor Comments__

    A detailed comparison of the sequence of the hydra PIP4K might help understand why it may not have worked like the sponge PIP4K. The discussion on the cnidarian PIP4K evolution is not convincing. It may not have worked because of it being expressed in a non-natural system. Structure prediction and comparison of proteins from different early branching animals should be used. Response: Thank you for these suggestions to understand why the cnidarian PIP4K may not have been functional. We will perform the suggested analysis and incorporate the data into the revision.

    78 - Multicellularity evolved many times. Maybe say 'first evolved metazoans'

    Response: Thank you for the suggestion.

    Line 598 A. queenslandica is not a coral, it's a sponge.

    Response: Text will be changed accordingly

    Line 612 'thcells' à 'the cells'

    Response: Text will be changed.

    Line 623 - full stop missing after metazoans.

    Response: Text will be changed

    Figure 1B - Classification should be consistent - C. elegans is a species name, whereas ctenophores and vertebrates belong to a different classification. Invertebrates is not a scientific group. The edges of the lines of the phylogenetic tree don't join and they need to be arranged correctly.

    Response: The names in the phylogeny will be changed to maintain uniformity. The representation of the phylogeny will be changed as mentioned.

    Figure 2B The full blot could be shown in the supplement.

    Response: Full blot will be provided as source data on resubmission or included as supplementary based on the destination journal’s specification.

    Optional

    Heterologous overexpression does not always provide the full picture of the gene functionality. To make claims on the evolution of function, testing gene functions homologous systems can give a better picture. For example, performing in vitro kinase activity assays of MbPIP4K after overexpressing PIP4K in Monosiga brevicollis. would be a great. Data is missing also about the presence and function of ctenophore PIP4K. Overexpression of ctenophore-PIP4K in Drosophila for functional analyses could help in understanding the distribution/diversity of function of PIP4K in early animals. Response: We agree with the reviewer that heterologous expression may sometimes not replicate the biochemical environment of cells in the organism from which the gene being expressed was originally derived. Yet, heterologous expression experiments do sometimes provide an insight into properties solely dependent on the polypeptide with limited or no contribution from the cellular environment. In principle expressing PIP4K in *M.brevicollis *cells and then performing kinase assays would be a very good idea. However, we would like to highlight that till date there has been only one study where septins have been transfected in Choanoflagellates and their localization being observed. We are not set up to culture *M. brevicollis *and will be unable to do this for a revision of the current manuscript. However, we appreciate the importance of this experiment and will do this in collaboration with a choanoflagellate lab in a follow up study to this one.

    Ctenophores like cnidarians have two main layers of cells that sandwich a middle layer of jelly-like material, while, more complex animals have three main cell layers and no intermediate jelly-like layer. Hence ctenophores and cnidarians have traditionally been labelled diploblastic. Studies have shown that ctenophores and unicellular eukaryotes share ancestral metazoan patterns of chromosomal arrangements, whereas sponges, bilaterians, and cnidarians share derived chromosomal rearrangements. Conserved syntenic characters unite sponges with bilaterians, cnidarians, and placozoans in a monophyletic clade while ctenophores are excluded from this clustering, placing ctenophores as the sister group to all other animals. Ctenophore PIP4K sequence can be identified and compared as discussed before to other PIP4K sequences used in this study.

    Reviewer #3 (Significance (Required)):

    Significance: This is the first study that addresses PIP signaling pathway in early metazoans. The findings of this manuscript contribute to the understanding of second-messenger signaling and its link with the origin and evolution of metazoan multicellularity. PIP signaling is crucial in different metazoan aspects such as cytoskeletal dynamics, neurotransmission, and vesicle trafficking, and hence, plays a critical role in metazoan multicellularity. Through this study, it was interesting to see that some components of the PIP signaling pathway are conserved in yeast, but some, such as the PIP4K protein evolved at the brink of metazoan evolution, highlighting the need for complexity in metazoans and their close relatives - the facultatively multicellular choanoflagellates. Since this is a crucial pathway in human biology and has medical significance due to its role in tumorigenesis and cancer cell migration, this study serves the audience in basic research such as evolutionary biology, and applied research such as human medicine. My field of expertise is molecular biology, cell biology and microbiology, with specific expertise on choanoflagellates. Therefore, it is exciting to see the homologs of PIP4K present in choanoflagellates.

    __ Evidence, Reproducibility, and clarity:__

    The authors have made a clear case of why PIP4K needs to be studied. They have thoroughly mapped PIP4K throughout the tree of life. The results are clear and reproducible. With the findings of this study, they have linked the PIP signalling cascade and metazoan evolution. Using the heterologous expression of sponge A. queensladica PIP4K, they have made compelling evidence that AqPIP4K functions in PIP5 phosphorylation, as seen in humans and Drosophila. However, it was not convincing why the hydra PIP4K was not functional. It was also not convincing why the PIP4K is metazoan-only when there is a conserved sequence (with conserved domain structure) present in choanoflagellates.

    We thank the reviewer for appreciating the novelty and importance of our findings in multiple areas of basic biology related to early metazoans and basic biomedical sciences. We also note their comments on the clear and reproducible results presented. Points raised related to the lack of functionality of PIP4K from Hydra and choanoflagellates are noted and will be addressed as indicated in response to other reviewer comments.

    Experiments/Analysis to be done

    1. We will perform a multiple sequence alignment using PIP4K sequences from multiple choanoflagellates and sponges to identify these differences.
    2. What we propose to do is to compare available PIP4K sequences from multiple Porifera and Cnidaria genomes and try and understand differences in the protein sequence that might explain differences in function.
    3. We will add more species of sponges, ctenophores, placozoans, and cnidarians in our analysis of PIP4K sequences. We will also include an analysis of other unicellular holozoans where genome sequence is available.
    4. We will perform the phylogenetic analysis of the phosphoinositide kinases in the format suggested by the reviewer and add it in the revision as a supporting evidence.
    5. Structure prediction and comparison of proteins from different early branching animals should be used.
    6. Uniformity of terminology and alignment with conventions in the field of animal taxonomy
    7. NCBI ID of sequences to be added and include more non-bilaterian animals sequences in phylogeny- redo the phylogeny.
    8. Check for PI signalling genes in choanoflagellates
    9. More detailed description of phylogenetic analysis.
    10. Add complete Western blot as source data.

    3. Description of the revisions that have already been incorporated in the transferred manuscript

    Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript. If no revisions have been carried out yet, please leave this section empty.

    4. Description of analyses that authors prefer not to carry out

    • Expression of PIP4K in choanoflagellates and in vitro kinase assays with lysates. It is beyond our technical ability to perform these experiments at this stage.
  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Summary

    In this manuscript, the authors investigate the evolutionary origins of metazoan Phosphatidylinositol phosphates (PIPs) signaling by elucidating the sequence and function of the PIP4K enzyme, which is crucial for converting PI5P to PI(4,5)P2 through phosphorylation. The authors have described PIP4K-like sequences distributed throughout metazoans and choanoflagellates through an extensive sequence screening. With in vitro and in vivo functional assays, the authors have shown that the sponge A. queenslandica PIP4K (AqPIP4K) is functionally similar to its human counterpart and highlight the major discovery of this study - that PIP4K protein function dates back to as early as sponges.

    Major comments

    There are two key limitations to this paper. Like the sponges, ctenophores are one of the earliest branching metazoans. They are not well addressed in the paper. Secondly, despite finding PIP4 homologs in choanoflagellates, the authors claim that PIP4 is metazoan-specific.

    1. Line 46: A. queenslandica is the earliest branching metazoan. The phylogeny of sponges and ctenophores is not conclusively defined and hence, the statement must be rephrased. Despite the brief description of the evolution of metazoan lineage in the discussion section, ctenophores are missing from the phylogenetic tree. At least a sequence-level information PIP4K in ctenophores would strongly back the claims of the manuscript. Here is the link to the Mnemiopsis database.
    2. Mentioning that choanoflagellates contain homologs of PIP4K contradicts the statement that PIP4K is metazoan-specific. As per Fig 1E., the domain organization of PIP4K is conserved among choanoflagellates and metazoans. What is the percent sequence similarity to the query? This could answer why it doesn't show activity in Drosophila rescues - the system might simply not be compatible with the choanoflagellate homolog. The same may apply to the cnidarian homolog HvPIP4K. Further evidence is needed before concluding that MbPIP4K doesn't phosphorylate PIP5. It is additionally fascinating that MbPIP4K localizes at the plasma membrane unlike other homologs - this function might be choano-specific. Overall, PIP4K's possible origin in the choanoflagellate-metazoan common ancestor backs the current research that choanoflagellates indeed hold clues to understanding metazoan evolution. Further research is necessary before concluding (as in line 648) in the discussions section, where it is mentioned that "PIP4K does not play any important functional role in choanos".

    Minor Comments

    1. A detailed comparison of the sequence of the hydra PIP4K might help understand why it may not have worked like the sponge PIP4K. The discussion on the cnidarian PIP4K evolution is not convincing. It may not have worked because of it being expressed in a non-natural system. Structure prediction and comparison of proteins from different early branching animals should be used.
    2. 78 - Multicellularity evolved many times. Maybe say 'first evolved metazoans'
    3. Line 598 A. queenslandica is not a coral, it's a sponge.
    4. Line 612 'thcells'  'the cells'
    5. Line 623 - full stop missing after metazoans.
    6. Figure 1B - Classification should be consistent - C. elegans is a species name, whereas ctenophores and vertebrates belong to a different classification. Invertebrates is not a scientific group. The edges of the lines of the phylogenetic tree don't join and they need to be arranged correctly.
    7. Figure 2B The full blot could be shown in the supplement.

    Optional

    1. Heterologous overexpression does not always provide the full picture of the gene functionality. To make claims on the evolution of function, testing gene functions homologously systems can give a better picture. For example, performing in vitro kinase activity assays of MbPIP4K after overexpressing PIP4K in Monosiga brevicollis. would be a great. Data is missing also about the presence and function of ctenophore PIP4K. Overexpression of ctenophore-PIP4K in Drosophila for functional analyses could help in understanding the distribution/diversity of function of PIP4K in early animals.

    Significance

    This is the first study that addresses PIP signaling pathway in early metazoans. The findings of this manuscript contribute to the understanding of second-messenger signaling and its link with the origin and evolution of metazoan multicellularity. PIP signaling is crucial in different metazoan aspects such as cytoskeletal dynamics, neurotransmission, and vesicle trafficking, and hence, plays a critical role in metazoan multicellularity. Through this study, it was interesting to see that some components of the PIP signaling pathway are conserved in yeast, but some, such as the PIP4K protein evolved at the brink of metazoan evolution, highlighting the need for complexity in metazoans and their close relatives - the facultatively multicellular choanoflagellates. Since this is a crucial pathway in human biology and has medical significance due to its role in tumorigenesis and cancer cell migration, this study serves the audience in basic research such as evolutionary biology, and applied research such as human medicine. My field of expertise is molecular biology, cell biology and microbiology, with specific expertise on choanoflagellates. Therefore, it is exciting to see the homologs of PIP4K present in choanoflagellates.

    Evidence, Reproducibility, and clarity:

    The authors have made a clear case of why PIP4K needs to be studied. They have thoroughly mapped PIP4K throughout the tree of life. The results are clear and reproducible. With the findings of this study, they have linked the PIP signalling cascade and metazoan evolution. Using the heterologous expression of sponge A. queensladica PIP4K, they have made compelling evidence that AqPIP4K functions in PIP5 phosphorylation, as seen in humans and Drosophila. However, it was not convincing why the hydra PIP4K was not functional. It was also not convincing why the PIP4K is metazoan-only when there is a conserved sequence (with conserved domain structure) present in choanoflagellates.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    The manuscript by Krishnan et al. uses molecular phylogenetics, in vitro kinase assays, heterologous expression assays in Drosophila S2 cells and mutant complementation assays in yeast to study the evolution and function of putative PIP4 kinase genes from a sponge, a cnidarian and a choanoflagellate. Based on these experiments, the authors conclude that PIP4K is metazoan-specific and that the sponge PIP4K has conserved functions in selectively phosphorylating PI5P.

    The study is in principle of interest and it could all be valid data, but the large number of flaws in the data presentation and/or analysis just makes it hard to assess the quality and thus validity of the data and conclusions.

    Major comments:

    Overall, the manuscript lacks scientific rigour in the analysis and representation of the results, and the validity of many of the conclusions is therefore difficult to assess.

    Major problems are:

    (i) The author base their study on the evolution of PIP4K genes on a deeply flawed concept of animal evolution. On multiple occassions, including the title, the authors refer to extant species (e.g. Amphimedon) as 'early metazoan', 'regarded as the earliest evolved metazoan' (l. 46-7) or 'the earliest examples of metazoans' just to name a few. This reflects a 'ladder-like' view on evolution that suggests that extant sponges are identical to early 'steps' of animal evolution. Also, the author's interpretation that one cluster of genes 'contained the sequences from early metazoans like sponges, cnidaria and nematodes' is referring to an outdated idea of animal phylogeny where nematodes were thought to be ancestrally simple organisms grouped as 'Acoelomata'. This idea of animal phylogeny was however disproven by molecular phylogenetics since the 1990ies.

    (ii) The description of taxa in the phylogenetic tree in Fig. 1B lacks any understanding of phylogenetic relationships between animals and other eukaryotic groups. What kind of taxa are 'invertebrates' or 'parasites'? And why would 'invertebrates' exclude cnidarians and sponges? Also, why is the outgroup of opisthokonts named 'Eukaryota'?? Are not all organisms represented on the tree eukaryotes?

    (iii) The methods part lacks any information about the type of analysis (ML, Bayesian, Parsimony?) used to perform the phylogenetic analysis shown in Fig. 1C. Also, the authors mention three distinct clusters (l.428) that are not labelled in the figure.

    (iv) The validity of the Western Blot is difficult to assess as the authors have cut away the MW markers. Without, it is for example difficult to assess the size differences visible between Hydra and Monosiga PIP4K-GFP proteins on Fig. 2B. Also, it has become standard practice to show the whole Western blot as supplementary data in order to assess the correct size of the bands and the specificity of the antibody. This is also missing from this manuscript.

    (v) The authors claim that AqPIP4K was able to convert PI3P into PI with very low efficiency (Figure 2E), but without further label in the figure or explanation, it remains unclear how the authors come to this conclusion.

    (vi) The box plots in Fig. 3C and D lack error bars and thus seem to be consisting of only single data points without replicates. Also, Fig. 3C is a quantification of Fig. 3B but it remains unclear what has been quantified and how. It is also unclear how %PIP2 was determined.

    (vii) Throughout Fig. 4, I do not understand the genotypes indicated on the x-axis of the plots and below the images. I read the figure legends and manuscript describing these results at least 3 times, but cannot figure out what it all means. On Fig. 4C, what is the wild-type situation?

    Significance

    If validated and put in the right phylogenetic context, the study is potentially contributing to expanding our knowledge on the evolution of metazoan-specific features, especially the evolution of proteins involved in cell-cell signalling and growth control.

    My field of expertise is broadly in evo-devo, molecular phylogentics, developmental genetics and cell biology. The in vitro lipid analysis seems interesting and potentially valid but I do not have sufficient expertise to evaluate its validity.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary:

    The researchers identified PIP4K (phosphatidylinositol 5 phosphate 4-kinase) as a lipid kinase that is specific to metazoans. In order to determine its conserved function across metazoans, they compared PIP4K activity in both early-branching metazoans and bilaterian animals. Biochemical assays demonstrated a conserved catalytic activity between the sponge Amphimedon queenslandica (AqPIP4K) and human PIP4K. In in-vivo experiments, AqPIP4K was found to rescue the reduced cell size, growth, and development phenotype in larvae of null mutant in Drosophila PIP4K. Based on these findings, the authors suggest that the function of PIP4K was established in early metazoans to facilitate intercellular communication. The experiments were well designed, and a range of biochemical, in vitro, and in vivo experiments were conducted. That being said, there are some questions that require further discussion before we can fully accept the author's conclusion of an evolutionarily conserved function of PIP4K across metazoans.

    Major comments:

    • The authors mentioned that PIP4K is metazoan-specific and involved in intercellular communication. How can we explain the presence of PIP4K in choanoflagellate genomes? Despite its high similarity with conserved domains and functionally important residues, experimental results with the PIP4K from Choanoflagellate (Monosiga brevicollis, MbPIP4K) such as Mass spectrometry-based kinase assay and mutant Drosophila PIP4K didn't show similar activity to sponge AqPIP4K. The authors suggested that "In the context of other ancient PIP4K it is possible that since choanoflagellates exist as both single-cell and a transient multicellular state and do not have the characteristics of metazoans, PIP4K does not play any important functional role in these." However, this explanation is not well justified; they need to provide a more detailed discussion on this.
    • Likewise, the PIP4K gene has been identified in cnidarians, which are a sister group to bilaterian animals. However, the Cnidaria HvPIP4K showed no activity in biochemical or functional assays. In comparison to sponges, cnidarians are relatively complex organisms, and I believe that PIP4K is highly important for intercellular communication, as it is in bilaterians. The authors attempted to explain this by suggesting that "Based on theories of parallel evolution between cnidarians and sponges during early metazoan evolution, it is possible that the PIP4K gene was retained functional in one lineage and not in other." However, I am not convinced by this statement.
    • Please provide details of the databases (Uniprot-KB, NCBI sequence database, Pfam) versions. After identifying the specific PIP4K protein in each species (e.g AqPIP4K and HvPIP4K), have you considered performing a reciprocal blast against the human genome to see if you have a top hit to PIP4K? Hence, the main focus of the project is on PIP4K as a metazoan-specific protein. We need to include a wider representation of non-bilaterian animals, including multiple species from sponges, ctenophores, placozoans, and cnidarians. Additionally, please check if homologues of PIP4K are present in other unicellular holozoans besides choanoflagellates.
    • Authors suggested the identification of other components of the PI signaling pathway along with PIP4k in the sponge. What is the status of these PI signaling pathway genes in other non-bilaterians and choanoflagellates?
    • phylogenetic tree of all PIP4K sequences (Figure 1C): How authors can be certain that the identified PIP4K sequences (e.g AqPIP4K, HvPIP4K, and MbPIP4K) are indeed PIP4K, especially when there are several closely related proteins? It is important to conduct phylogenetic analysis alongside other PIP sequences (such as PI3K, PI4K, PIP5K, and PIP4K). If this analysis is carried out, the identified AqPIP4K, HvPIP4K, and MbPIP4K should be grouped together with human PIP4K in the same cluster.

    Minor comments:

    • Line 157: Phylogenetic conservation of PIP4Ks: Please provide details about bootstrap analysis.
    • Line 230: symbol correction 30{degree sign}C
    • Line 429-430: "from early metazoans like Sponges, Cnidaria and Nematodes." Nematodes are not considered early metazoans.
    • Line 477-478: "However, interestingly, MbPIP4K::GFP localizes only at the plasma membrane in S2 cells (Figure 2C)." This part was not further discussed. Can you please elaborate on why MbPIP4K::GFP localizes only at the plasma membrane in S2 cells?
    • Line 598: "the earliest examples of metazoa, namely the coral A.queenslandica" A.queenslandica is a sponge, not coral.
    • Line 602: "Amphimedon and human enzyme, although separated by 50Mya years of evolution" I think it's 500 million years ago, not 50 million years ago.
    • Line 612: "co-ordinated communication between thcells is the most likely function" the cell.
    • Line 614: "intracellular phosphoinositide signalling The identity of the hormone" missing full stop punctuation.
    • Line 802 - 804: "other by way of difference in colour. The sub clusters have been numbered (1- early metazoans, 2- Nematodes, 3- Arthropods, 4- Molluscs, 5- Vertebrates (isoform PIP4K2C), 6- Vertebrates (isoform PIP4K2A), 7- Vertebrates (isoform PIP4K2B)." In the Figure, I can't find numbers on the subclusters.
    • Line 805- 807: "Phylogenetic analysis of selected PIP4K sequences from model organisms of interest. PIP4K from A.queenslandica has been marked in rectangular box." The rectangular box is missing in the figure.
    • Figure 1C: full forms of species names are missing.

    Significance

    The data is presented well, and the authors used a wide range of assays to support their conclusion. The study is highly impactful and can have a broader influence on the scientific community, particularly in evolutionary molecular biology, development, and biochemistry.

    The study provides interesting findings; however, the reasons for PIP4K not being functional in cnidarians as in sponges and why PIP4K is present in unicellular holozoans but not functional are unclear.

  5. Version published to 10.1101/2024.08.01.606031v1 on bioRxiv