The Fraser complex interconnects tissue layers to support basal epidermis and osteoblast integrated morphogenesis underlying fin skeletal patterning

  1. Amy E. Robbins
  2. Samuel G. Horst
  3. Victor M. Lewis
  4. Scott Stewart
  5. Kryn Stankunas

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Fraser Syndrome is a rare, multisystemic autosomal recessive disorder characterized by disrupted epithelial-mesenchymal associations upon loss of Fraser Complex genes. Disease manifestation and affected organs are highly variable. Digit malformations such as syndactyly are common but of unclear developmental origins. We explored if zebrafish fraser extracellular matrix complex subunit 1 (fras1) mutants model Fraser Syndrome-associated appendicular skeleton patterning defects. Approximately 10% of fras1 mutants survive to adulthood, displaying striking and varied fin abnormalities, including endochondral bone fusions, ectopic cartilage, and disrupted caudal fin symmetry. The fins of surviving fras1 mutants frequently have fewer and unbranched bony rays. fras1 mutant fins regenerate to their original size but with exacerbated ray branching and fin symmetry defects. Single cell RNA-Seq analysis, in situ hybridizations, and antibody staining show specific Fraser complex expression in the basal epidermis during regenerative outgrowth. Fras1 and Fraser Complex component Frem2 accumulate along the basal side of distal-most basal epidermal cells. Greatly reduced and mislocalized Frem2 accompanies loss of Fras1 in fras1 mutants. The Sonic hedgehog signaling between distal basal epidermis and adjacent mesenchymal pre-osteoblasts that promotes ray branching persists upon Fraser Complex loss. However, fras1 mutant regenerating fins exhibit extensive sub-epidermal blistering associated with a disorganized basal epidermis and adjacent pre-osteoblasts. We propose Fraser Complex-supported tissue layer adhesion enables robust integrated tissue morphogenesis involving the basal epidermis and osteoblasts. Further, we establish zebrafish fin development and regeneration as an accessible model to explore mechanisms of Fraser Syndrome-associated digit defects and Fraser Complex function at epithelial-mesenchymal interfaces.

Article activity feed

  5. Review Commons

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

    Learn more at Review Commons

    Reply to the reviewers

    The authors do not wish to provide a response at this time.

  6. 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:

    This paper provides an elegant investigation into the dysmorphology of skeletal structures upon loss of the extracellular Fraser Complex through use of a zebrafish fras1 mutant. The adult phenotypes of this mutant had not been previously explored in detail. The authors use transgenics, histological staining and immunostaining to visualize the morphological defects, then examine Fraser Complex localization and effects on Shh signaling upon loss of Fras1. They analyse both steady state and regeneration contexts. Most importantly, they demonstrate the blistering that has been shown to occur in larval fins, also occurs during regeneration of the adult fins, and that there is a disorganization of the pre-osteoblasts due to basement membrane corruption. This work provides novel investigation into how Fraser Complex leads to skeletal defects.

    Major comments:

    Overall the data presented by the authors is logical and very well presented. They make reasonable claims that are supported by sound experimentation. Statistics are used appropriately and the authors combine different approaches to make their points. For example they draw on previous single cell RNA-seq Data sets to define the cell type expressing Fraser complex components but then also use immunostaining and ins situ hybridisation to localise expression domains. I thought the analysis using the ptch1:kaede line to be particularly elegant and informative.

    My first major question relates to Figure 7. The authors cage their analysis under the impetus of understanding how the distal anomalies and skeletal defects in Fraser Syndrome arise in development. They present convincing images of distal blistering during regeneration. Were similar blisters seen at any stage during adult fin development prior to the full fin formation? The authors might have noted this during their imaging of the ptch1 reporter Figure 6 Supplement 1. Or they might need to look earlier. Figure 7 is informative analysis. It's a shame to limit it to regeneration and not look during adult fin formation which would have direct inferences for human development.

    Secondly, was osteoblast morphology affected as well as their patterning in the fras1 mutants? Perhaps the authors could look at zns-5 antibodies or sp7:egfp line in transverse cryosections to assess if there was a change in osteoblast morphology. Figure 5B' certainly suggests they might be more cuboidal and have lost their flattened shape. This change in osteoblast morphology has been noted in other ECM mutants affecting the lepidotrichia.

    Minor comments:

    I have only a few minor comments.
    Line 55 Introduction. For clarity change this sentence to "variable expressivity and incomplete penetrance reminiscent of the variability seen in distal limb defects in humans with Fraser Syndrome."

    Referring to Figure 2, is each fin ray thicker in fras1 mutants than in WT? It appears so but might be just an optical illusion. Would be easy to measure and state in the text.

    Line 242, 243 and Figure 5. The text refers to Fig 5C' and Fig 5D', but I could only find Fig 5C' and Fig 6C'. Do you mean E, F??

    Related you claim that in fras1 mutants, frem2 protein remains intracellular. How do you know that protein signal is intracellular yet in the WT it is extracellular? Please reword this or show more conclusively. This is also repeated on line 360 in the discussion.

    Fig 6B, D are not referred to in the main text.

    I couldn't find details of the Runx2 antibody in the materials and methods

    Significance

    This paper provides novel insights into how loss of Fraser Complex might alter morphology of post-embryonic structures, and gives novel, visual indication of the effect basement membrane disruption has on osteoblast patterning. This will give novel guidance to explaining the basis of skeletal presentations of Fraser Syndrome for basic researchers interested in the importance of osteoblast environment on patterning and clinicians examining similar rare skeletal congenital syndromes.

    It elegantly demonstrates that cellular topology, organisation and morphology is just as important as signalling in defining organ growth and shape. The authors also suggest this model of an indirect disruption of osteoblast patterning might explain why there is such variability in distal skeletal phenotype severity in Fraser patients. This is a valid and reasonable hypothesis.

    My background directly relates to mutant analysis of zebrafish fins

  7. 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

    Summary:

    In this manuscript, Robbins et al. develop the adult zebrafish caudal fin as a model for limb deformities in Fraser syndrome. They show that the zebrafish fras1 mutant can survive to adulthood and investigate how loss of fras1 influences the caudal fin during its regeneration. The fin ray branching defects and chondral bone defects examined in this study show a degree of variability akin to many defects found in Fraser syndrome, including limb defects. The paper is an easy read, has beautiful imaging, and shows compelling quantification. As such, the authors set up the zebrafish fras1 mutant caudal fin as a novel and interesting platform to investigate the onset and variability of limb defects in Fraser syndrome.

    Major comments:

    Figure 7 makes some critical claims based solely on H&E staining; better markers are needed. I also strongly recommend that the authors repeat this Figure 7 experiment with immunolabel or other techniques that mark the epithelial and mesenchymal layers distinctly. In particular, since the author's discussion focuses on osteoblast positioning, it seems important to mark the osteoblasts alongside the epithelium.

    The authors posit that background mutations explain Fraser syndrome variation, but it's worth also considering non-genomic origins of variation, including potential micro-environmental differences and/or stochasticism. The caudal fin seems like an ideal system to test the idea - one could injure the fin and test how well initial defects predict defects after regeneration. It looks like the authors started to do so in panel 3R, but don't comment on the finding. Also, it looks like the initial severity doesn't completely predict regenerated severity - for instance, the least severe fin pre-injury becomes the most severe fin post-injury. I recommend explaining panel 3R in results text and returning to it in the discussion, to contextualize how this finding might influence understanding of fras1-dependent variability. OPTIONAL: Variability could also be compared between different fins in the same animal; does the severity of caudal fin defect correlate with severity in the pelvic or dorsal fin? If so or if not, what might this suggest about the origins of phenotypic variation in this model?

    The study is focused on core Fraser complex, particularly fras1 and the Frem genes which are mutually-stabilizing components of anchoring cords that link nascent epithelia to underlying mesenchyme. The authors could broaden the paper's impact and scope by considering other proteins bound by the core Fraser complex, such as AMACO, Integrins, Npnt, fibrillin, etc. Likewise, examination of Collagen expression, such as collagen VII, may help explain why there is a dissipation over time for the requirement for fras1 to prevent blistering. It may be outside the scope of this study to delve deep into these 'nearby genes' but it seems reasonable to examine them bioinformatically in the scRNA-seq (perhaps adding a supplemental figure if the results are unsatisfying), or at least to discuss them and thereby contextualize the overall findings.

    Minor comments:

    Figure 8 feels like a visual abstract, summarizing findings and contextualizing existing models to the caudal fin, instead of putting forward a truly new model. It focuses on the concept that Fras1 is needed for epithelial-mesenchymal attachments and that Fraser-components stabilize one another; both of these ideas are over a decade old. Although the concepts are cited appropriately within the discussion narrative, it would be nice if Figure 8 did a bit more. The model could be revised in a way that offers new insights into how the specific defects seen in this study arise, by hinting at answers to some of the many questions raised by the study. What is the source of variation? Why do the fin rays fail to branch in the mutant? (could there be more to that decision than simple osteoblast mispositioning?) Why does defect severity change during regeneration vs. development? Why does an extra hypural cartilage form in the mutant? Is there any similarity between the failure to branch fin rays and the presence of fusions in chondral skeleton? These fusions could also be compared and contrasted with non-limb skeletal fusions? It's certainly optional to tackle all of these issues, but discussing any of them could increase the manuscript's impact and establish interesting fodder for future papers. These changes are important, but I place this remark in 'minor comments', because an improved final figure would not be critical to publication in some journals.

    This sentence on line 273-274 is confusing and should be revised: "However, the Fraser Complex at least supports the Shh/Smo downstream cell behaviors that split pre-osteoblasts given frequent ray branching defects in fras1-/- mutants." What do the authors mean by "at least supports"? The phrasing may imply 'Provides physical support essential to,' but that reading does not explain why the fin rays branch in Shh expressing cells regardless of the presence of Fras1. (this comment is actually 'minor')

    The Figure 7 title states that the Fraser complex is needed for normal "epithelial-mesenchymal tissue layering." I am not certain what the authors mean by this. The phrase written in the figure title implies that mesenchymal cells start appearing inside of epithelial regions, but that interpretation doesn't match the figure nor results text. An alternative read - one consistent with the final figure - is that the authors are simply trying to show that the blisters form at the interface of the fin and underlying mesenchyme. If this is the primary thrust of the argument, then it could be stated plainly - and shown more clearly in the results section. (this comment is also truly 'minor')

    Significance

    This manuscript establishes the zebrafish adult caudal fin as a new model to investigate epithelial-mesenchymal interaction defects and skeletal variation caused by loss of the central Fraser complex gene. It is an important contribution to existing models because of the similarity between patterning mechanisms in the caudal fin and tetrapod limbs, which are variably disrupted in Fraser's syndrome. This variability is itself of interest, because the profound variability of Fraser phenotype offers an experimental model for high-variance diseases, which remain poorly understood. The caudal fin is an exciting system to study variation, in particular because it can be cut and regrown rapidly with phenotypic severity compared between regenerates in a clonal system; that regenerative tractability is particularly potent when paired with the zebrafish transgenic toolbox, as the authors show in Figure 6. The present manuscript feels almost complete and yet it opens up many questions, suggesting that it can serve as a good foundation for future studies.

  8. 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

    Robbins and colleagues present a new zebrafish model for studying the rare disorder Fraser Syndrome in a well written manuscript. The authors present an interesting phenotype caused by a mutation in fras1: mutants have small, misshapen fins with rays that frequently fail to form branches. The authors show that the Fras1 protein localizes to the distal portions of regenerating fin rays and that in its absence, another component of the Fraser Complex (Frem2) is decreased and mislocalized. The authors show that Fras1 is not required for Shh/Smo signal transduction but that branching is nonetheless disrupted.

    Major comments:

    Since the basement membrane ends up being such a big part of the story, it would add support to the model to specifically present a laminin stain in both WT and the mutant. This could potentially support to the idea that the FPC is integrating the BM with the osteoblasts and it would be helpful to see where the BM is relative to the 'blisters.'

    The fras1-/- mutants show a near-absence of Frem2 protein; the authors conclude that Fras1 supports Frem2 secretion and assembly into Fraser Complex-containing BM. However have the authors considered the possibility that Fras1 directly or indirectly regulates Frem2 transcription? (The model in Fig 8 seems to show a relative increase in Frem2, which should be altered)
    Do the other components of the FPC (Frem1a/b, Frem3) show similarly disrupted levels and localization? As presented, since the Frem1a/b, Frem3 are not actually examined in a fras1-/- context, it is not justified to show their intracellular localization in the model in Fig 8b.

    The relative activities of osteoblasts and osteoclasts may regulate the relative location of branches (Cardeira-da-Silva et al 2022). In the fras1 system mutant, it would be beneficial to examine osteoclasts to rule out the idea that Fras1 is simply required for osteoclast activity. This could additional lend support to the idea that disrupted integration between the BE and Obs underlies that failure of fras1-/- mutants to form branches. This could be done within a few months by crossing the mutant into an osteoclast reporter, or more rapidly by using an osteoclast specific antibody or ISH.

    The authors look only at the meristic presence/absence of fin ray branches, and do not take fin length into account. Longer rays are more likely to form branches and the fras1-/- have shorter fins. Some of the branches may be absent in the mutant background simply because the rays are not long enough to have formed them. How far are the branches from the body in each background? Where do the branches form along the total length of the rays?
    In the regeneration experiments, do the mutants regenerate at the same relative rate as WTs? Could the decrease in the number of branched rays simply be due to the fact that the mutants may not have not regenerated to their original length by 28 dpa? Has the distance from the body to branch changed? These are all addressable with the data the authors have already collected.

    Minor comments:

    The protruding lower jaw phenotype mentioned in the results should either be shown, or if it is redundant with previous publications please add the citation here.

    Please show Amputation planes in Fig 4E-G. It is notable that Fras1 seems to be present in the proximal region (proximal to the amputation plane?). This seems like it deserves mention in the manuscript.

    In line 195, please clarify what you mean by ray proportions. The proportion of longest/shortest, presumably?

    Please change the title of Fig 3: Fras1 mutants robustly restore skeletal patterning. Also, the ray length ratio in Fig 3Q has been tested with an unpaired t test. It should be a paired test as in R.

    This is an unbelievably a minor point, but for consistency with the other panels I think that in Fig 5 E and F should be C' and D'.

    The titles of Fig 6 and Fig 7 should indicate indicate Fras1 not the Fraser Complex. Eg. Fras1 is not essential for sonic hedgehog/smoothed signal transduction during fin regeneration

    There are also a few grammar errors that should be fixed.

    Significance

    Robbins and colleagues present a new zebrafish model for studying the rare disorder Fraser Syndrome.

  9. Version published to 10.1101/2023.07.08.548238v1 on bioRxiv