Higher vertebrates acquire a definitive body size around the time of their sexual maturation. Although many adult stem cells (aSCs) remain active and keep producing new cells afterwards, they mainly replace cells that are lost on a daily basis. On the other hand, lower vertebrates like fish keep increasing their size even during adulthood due to the capacity of aSCs to drive growth in parallel to maintaining organ homeostasis. The basis for the different outputs between aSCs in lower and higher vertebrates is still not fully understood. It has been reported, however, that in pathological conditions mammalian aSCs exhibit the ability to drive growth, as best represented by cancer stem cells (CSCs) (Batlle and Clevers, 2017; Nassar and Blanpain, 2016; Clevers, 2011; Suvà et al., 2014; Quintana et al., 2008; Barker et al., 2009; Schepers et al., 2012; Boumahdi et al., 2014).

Since stem cells in fish maintain homeostasis and drive post-embryonic growth in a highly controlled manner, the system permits identifying similarities and differences in case both functions are performed by dedicated populations or identifying specific conditions within a common stem cell pool driving homeostasis and growth. There are several genetic tools and techniques to explore aSCs in fish, and an abundant literature covering different aspects of their biology in various organs and also during regeneration paradigms (Gupta and Poss, 2012; Knopf et al., 2011; Tu and Johnson, 2011; Kizil et al., 2012; Kyritsis et al., 2012; Pan et al., 2013; Centanin et al., 2014; Jungke et al., 2015; Henninger et al., 2017; Singh et al., 2017; McKenna et al., 2016; Aghaallaei et al., 2016). Despite all these major advances, we still do not understand whether the same pool of stem cells is responsible for driving both growth and homeostatic replacement, or if alternatively, each task is performed by dedicated aSCs.

We decided to address this question using the medaka gill, which works as a respiratory, sensory and osmoregulatory organ in most teleost fish. Gills are permanently exposed to circulating water and therefore have a high turnover rate (Chrétien and Pisam, 1986). Additionally, their growth pace must guarantee oxygen supply to meet the energetic demands of growing organismal size. Moving from the highest-level structure to the smallest, gills are organised in four pairs of branchial arches, a number which remains constant through the fish’s life. Each brachial arch consists of two rows of an ever-increasing number of filaments that are added life-long at both extremes (Figure 1A). Primary filaments have a core from which secondary filaments, or lamellae, protrude. The lamellae are the respiratory units of the organ, and new lamellae are continually produced within each filament (Wilson and Laurent, 2002). Bigger fish, therefore, display more filaments that are longer than those of smaller fish, and there is a direct correlation of filament length and number and the body size of the fish (Wilson and Laurent, 2002).

Figure 1

Download asset Open asset

Besides being the respiratory organ of fish, the gill has additional functions as a sensory and osmoregulatory organ (Sundin and Nilsson, 2002; Wilson and Laurent, 2002; Jonz and Nurse, 2005; Hockman et al., 2017). It contains oxygen-sensing cells (Jonz et al., 2004), similar to those found in the mammalian carotid body although with a different lineage history (Hockman et al., 2017), and mitochondrial rich cells (MRCs) (Wilson and Laurent, 2002) that regulate ion uptake and excretion and are identified by a distinctive Na+, K+, ATPase activity. Other cell types include pavement cells (respiratory cells of the gills), pillar cells (structural support for lamellae), globe cells (mucous secretory cells), chondrocytes (skeleton of the filaments) and vascular cells. All these cell types must be permanently produced in a coordinated manner during the post-embryonic life of fish. The gill constitutes, therefore, an organ that allows addressing adult stem cells during the addition and homeostatic replacement of numerous, diverse cell types.

Bona fide stem cells can only be identified and characterised by following their offspring for long periods to prove self-renewal, the defining feature of stem cells (Clevers and Watt, 2018). In this study, we use a lineage analysis approach that revealed growth and homeostatic stem cells in the medaka gill. We found that gill stem cells are fate-restricted, and identified at least four different lineages along each filament. By generating clones at different stages, we show that these four lineages are generated early in embryogenesis, previous to the formation of the gill. Our results also indicate that growth and homeostatic aSCs locate to different regions along the gill filaments and the branchial arches. Homeostatic stem cells have a fixed position embedded in the tissue and generate cells that move away to be integrated into an already functional unit, similarly to mammalian aSCs in the intestinal crypt (Barker et al., 2008). Growth stem cells, on the other hand, locate to the growing edge of filaments and are moved as filaments grow, resembling the activity of plant growth stem cells at the apical meristems (Greb and Lohmann, 2016). We have also found that the homeostatic aSCs can turn into growth aSCs when the apical part of a filament is ablated, revealing that the activity of a stem cell is highly plastic and depends on the local environment. Our data reveal a topological difference between growth and homeostatic stem cells, which has similar functional consequences in diverse stem cell systems.

In this study, we use genetic lineage analysis and mathematical modelling to reveal the rationale behind the permanent post-embryonic growth in a vertebrate organ. We introduce the fish gill, and particularly branchial arches, as a new model system that displays an exquisite temporal/spatial organisation, and use it to characterise growth and homeostatic stem cells. We reveal two domains harbouring growth stem cells: both extremes of each branchial arch contain br-archSCs, which in turn generate filamSCs that locate to the tip of newly formed filaments. Additionally, filamSCs generate homeostatic stem cells at the lamellae along the longitudinal axis of the filament. The peripheral-to-central axis of branchial arches reflects a young-to-old filament order, and the longitudinal axis of a filament reflects a young-to-old lamellae order. The two growth stem cells and the one homeostatic stem cell types are clonal and organised in a hierarchical manner.

Our observations indicate that the relative position within the organ has a major impact on the growth vs homeostatic activity of stem cells. We have found that when the growth domain of a filament is lost, the homeostatic domain is able to generate a new, functional growth domain. This observation suggests that physical or molecular modifications in the local environment (relaxation of the inner core, or the absence of a repressive signal, respectively) could convert homeostatic stem cells into growth stem cells. In the absence of specific markers to label homeostatic stem cells before the ablation, however, we cannot discard the presence of quiescent stem cells that get activated after injury, nor the possibility of injury-triggered trans-differentiation as shown in the zebrafish caudal fin (Knopf et al., 2011).

Permanent post-embryonic growth is a challenging feature for an organism, since new cells have to be incorporated into a functional organ without affecting its physiological activity. Restricting growth stem cells to the growing edge is an effective way to compartmentalise cell addition and organ function. Strikingly, the location of growth stem cells in gill filaments is highly reminiscent of the overall topology of meristems in plants (Greb and Lohmann, 2016). In both systems, axis extension occurs by the sustained activity of stem cells that locate to the growing edge. These stem cells consistently remain at the growing zone, while their progeny start differentiation programs and occupy a final location at the coordinates in which they were born. It is to note that other ever-growing organs in fish follow the same growing principle, with tissue stem cells located at the growing edge and differentiated progeny left behind, as it has been nicely shown for different cell types in the zebrafish caudal fin (Tu and Johnson, 2011) and the medaka neural retina and retinal epithelium (Centanin et al., 2011; Centanin et al., 2014). Since stem cells are thought to have evolved independently in the vegetal and the animal lineages (Meyerowitz, 2002; Scheres, 2007), our results illustrate how the same rationale to sustain permanent growth can be adopted in the most diverse systems.

We have performed an organ-scale lineage analysis at cellular resolution and found that growth stem cells and homeostatic stem cells are fate-restricted. We used diverse un-biased labelling approaches (sub-optimal tamoxifen concentrations using an ubiquitously-expressed inducible ^ErT2CRE, heat-shock-induced expression of CRE, transplantation of few permanently-labelled cells at the blastula stage) to identify at least four different fate-restricted gill stem cells, which generate reproducible labelling patterns along gill filaments. Our experiments indicate that these stem cells do not share a common early progenitor, but rather have independent embryonic origins. Since each filament contains all four fate-restricted stem cells (we have not observed filaments lacking one entire lineage), our results determine that the growth zone of a gill filament is indeed an ensemble — a group of stem cells with different potencies that work in an interconnected manner. Two relevant avenues open from this analysis, namely: a) how stem cells are recruited together to a newly forming filament, a process that happens hundreds of times during the lifetime of a medaka fish and thousands of times in longer-lived teleost fish, and b) how stem cells coordinate their activity to maintain the ratio of cell types in the individual filaments to guarantee its proper growth. One fundamental aspect to start addressing coordination is to define the number of stem cells for each lineage, a parameter that proved to be hard to estimate for most vertebrate organs. The prediction for gill filaments is that they contained a very reduced number of stem cells, for they generate all-or-none labelled filaments of a given cell type reflecting a clonal nature.

We exploited the fate-restriction of growth stem cells in the gill system by generating chimeras that contain wild-type and a p53 mutant cells, and focused on filaments that display one lineage coming from donor cells and the other three coming from host cells. Our experiments revealed that P53 is indeed necessary to establish/maintain the coordinated growth of neighbour filaments. Strikingly, we observed growth phenotypes only when the lineage of pillar cells was involved: p53 mutant filaments that contain wild-type pillar cells can not growth at the same pace than fully p53 mutant filaments in the same branchial arch. Wild-type cells in other lineages did not produce the same phenotype, which we interpret as a differential hierarchical organisation of the different fate-restricted stem cells sustaining filament growth. Additionally, our experiments working with genetic chimeras and also the ones in which we ablate the growth domain of wild type filaments show that filament growth is a autonomous - that is a filament's growth rate does not adjust to the size of its neighbour filaments. This observation is particularly intriguing considering the order in filament size along a branchial arch, where every filament is larger that its younger neighbour and shorter than its older neighbour. Altogether, our results position the fish gill as an ideal system to quantitatively explore stem cell niches hosting multiple lineage-restricted stem cells, particularly the growth coordination within and between functional units.

In most adult mammalian organs, stem cells maintain homeostasis by generating new cells that will replace those lost during physiological or pathological conditions. We have functionally identified homeostatic stem cells in the fish gill, and focussed on the ones generating pillar cells. Our lineage analysis demonstrates that growth and homeostatic stem cells are clonal along a filament, where the former generate the latter. The most obvious difference between these two stem cell types is their relative position; growth stem cells are located at the growing tip, beyond the rigid core that physically sustains the structure of the filament, while homeostatic stem cells are embedded inside the tissue, adjacent to the collagen-rich chondrocyte column. It is to note that both the function and the relative location of the gill homeostatic stem cells match those of the mammalian homeostatic stem cells, being located at a fixed position and displacing their progeny far away - as it is observed for intestinal stem cells, skin stem cells and oesophagus stem cells among others (Barker et al., 2008; Blanpain and Fuchs, 2009; Seery, 2002). The functional comparison of growth and homeostatic stem cells in the gill suggests the existence of a physical niche that would restrict stem cells to their homeostatic role, preventing them to drive growth. Indeed, our RNA-seq data comparing the transcriptomes from the growth domain and the central domains revealed numerous Integrins and membrane proteins overrepresented at the growing edge, supporting the notion of a different physical niche mediating filament growth.

We believe that during vertebrate evolution, the transition from lower (ever-growing) to higher (size-fixed) vertebrates involved restraining the growth activity of adult stem cells. One of the main functions of mammalian physical niches, in this view, would be to restrict stem cells to their homeostatic function. Many stem-cell-related pathological conditions in mammals involve changes in the microenvironment including physical aspects of the niche (Brabletz et al., 2001; Vermeulen et al., 2010; Ye et al., 2015; Oskarsson et al., 2011; Liu et al., 2012; Butcher et al., 2009), suggesting that homeostatic stem cells could drive growth in that context. Along the same line, the extensive work using organoids that are generated from adult homeostatic stem cells, like intestinal stem cells, (Sato et al., 2009; Kretzschmar and Clevers, 2016), demonstrates that healthy aSCs have indeed the capacity to drive growth under experimental conditions and when removed from their physiological niche. Our work, therefore, illustrates how different niches affect the functional output of clonal stem cells driving growth and homeostatic replacement in an intact in vivo model.

All data analysed for this study is included in the manuscript and supporting files. Raw sequencing data have been deposited in GEO under accession code GSE130939.

1. 1. Buono L
   2. Martinez-Morales JR
   3. Centanin L

   (2019) NCBI Gene Expression Omnibus

   ID GSE130939. Transcriptome analysts of a growth domain and a differentiated domain of the medaka gill.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE130939

1. 1. Aghaallaei N
   2. Gruhl F
   3. Schaefer CQ
   4. Wernet T
   5. Weinhardt V
   6. Centanin L
   7. Loosli F
   8. Baumbach T
   9. Wittbrodt J

   (2016) Identification, visualization and clonal analysis of intestinal stem cells in fish

   Development 143:3470–3480.

   https://doi.org/10.1242/dev.134098

   • PubMed
   • Google Scholar
2. 1. Barker N
   2. van de Wetering M
   3. Clevers H

   (2008) The intestinal stem cell

   Genes & Development 22:1856–1864.

   https://doi.org/10.1101/gad.1674008

   • PubMed
   • Google Scholar
3. 1. Barker N
   2. Ridgway RA
   3. van Es JH
   4. van de Wetering M
   5. Begthel H
   6. van den Born M
   7. Danenberg E
   8. Clarke AR
   9. Sansom OJ
   10. Clevers H

   (2009) Crypt stem cells as the cells-of-origin of intestinal cancer

   Nature 457:608–611.

   https://doi.org/10.1038/nature07602

   • PubMed
   • Google Scholar
4. 1. Batlle E
   2. Clevers H

   (2017) Cancer stem cells revisited

   Nature Medicine 23:1124–1134.

   https://doi.org/10.1038/nm.4409

   • PubMed
   • Google Scholar
5. 1. Blanpain C
   2. Fuchs E

   (2009) Epidermal homeostasis: a balancing act of stem cells in the skin

   Nature Reviews Molecular Cell Biology 10:207–217.

   https://doi.org/10.1038/nrm2636

   • PubMed
   • Google Scholar
6. 1. Boumahdi S
   2. Driessens G
   3. Lapouge G
   4. Rorive S
   5. Nassar D
   6. Le Mercier M
   7. Delatte B
   8. Caauwe A
   9. Lenglez S
   10. Nkusi E
   11. Brohée S
   12. Salmon I
   13. Dubois C
   14. del Marmol V
   15. Fuks F
   16. Beck B
   17. Blanpain C

   (2014) SOX2 controls tumour initiation and cancer stem-cell functions in squamous-cell carcinoma

   Nature 511:246–250.

   https://doi.org/10.1038/nature13305

   • PubMed
   • Google Scholar
7. 1. Brabletz T
   2. Jung A
   3. Reu S
   4. Porzner M
   5. Hlubek F
   6. Kunz-Schughart LA
   7. Knuechel R
   8. Kirchner T

   (2001) Variable beta-catenin expression in colorectal cancers indicates tumor progression driven by the tumor environment

   PNAS 98:10356–10361.

   https://doi.org/10.1073/pnas.171610498

   • PubMed
   • Google Scholar
8. 1. Butcher DT
   2. Alliston T
   3. Weaver VM

   (2009) A tense situation: forcing tumour progression

   Nature Reviews Cancer 9:108–122.

   https://doi.org/10.1038/nrc2544

   • PubMed
   • Google Scholar
9. 1. Centanin L
   2. Hoeckendorf B
   3. Wittbrodt J

   (2011) Fate restriction and multipotency in retinal stem cells

   Cell Stem Cell 9:553–562.

   https://doi.org/10.1016/j.stem.2011.11.004

   • PubMed
   • Google Scholar
10. 1. Centanin L
    2. Ander JJ
    3. Hoeckendorf B
    4. Lust K
    5. Kellner T
    6. Kraemer I
    7. Urbany C
    8. Hasel E
    9. Harris WA
    10. Simons BD
    11. Wittbrodt J

    (2014) Exclusive multipotency and preferential asymmetric divisions in post-embryonic neural stem cells of the fish retina

    Development 141:3472–3482.

    https://doi.org/10.1242/dev.109892

    • PubMed
    • Google Scholar
11. 1. Chrétien M
    2. Pisam M

    (1986)

    Cell renewal and differentiation in the gill epithelium of fresh- or salt-water-adapted euryhaline fish as revealed by [3H]-thymidine radioautography [lebistes reticulatus]

    Biology of the Cell 56:137–150.

    • Google Scholar
12. 1. Clevers H

    (2011) The cancer stem cell: premises, promises and challenges

    Nature Medicine 17:313–319.

    https://doi.org/10.1038/nm.2304

    • PubMed
    • Google Scholar
13. 1. Clevers H
    2. Watt FM

    (2018) Defining Adult Stem Cells by Function, not by Phenotype

    Annual Review of Biochemistry 87:1015–1027.

    https://doi.org/10.1146/annurev-biochem-062917-012341

    • PubMed
    • Google Scholar
14. Software

    1. Goff L
    2. Trapnell C
    3. Kelley DR

    (2018)

    cummeRbund: Analysis, Exploration, Manipulation, and Visualization of Cufflinks High-Throughput Sequencing Data, version R Package Version 2.24.0

    cummeRbund: Analysis, Exploration, Manipulation, and Visualization of Cufflinks High-Throughput Sequencing Data.
15. 1. Greb T
    2. Lohmann JU

    (2016) Plant stem cells

    Current Biology 26:R816–R821.

    https://doi.org/10.1016/j.cub.2016.07.070

    • PubMed
    • Google Scholar
16. 1. Gupta V
    2. Poss KD

    (2012) Clonally dominant cardiomyocytes direct heart morphogenesis

    Nature 484:479–484.

    https://doi.org/10.1038/nature11045

    • PubMed
    • Google Scholar
17. 1. Henninger J
    2. Santoso B
    3. Hans S
    4. Durand E
    5. Moore J
    6. Mosimann C
    7. Brand M
    8. Traver D
    9. Zon L

    (2017) Clonal fate mapping quantifies the number of haematopoietic stem cells that arise during development

    Nature Cell Biology 19:17–27.

    https://doi.org/10.1038/ncb3444

    • PubMed
    • Google Scholar
18. 1. Hockman D
    2. Burns AJ
    3. Schlosser G
    4. Gates KP
    5. Jevans B
    6. Mongera A
    7. Fisher S
    8. Unlu G
    9. Knapik EW
    10. Kaufman CK
    11. Mosimann C
    12. Zon LI
    13. Lancman JJ
    14. Dong PDS
    15. Lickert H
    16. Tucker AS
    17. Baker CV

    (2017) Evolution of the hypoxia-sensitive cells involved in amniote respiratory reflexes

    eLife 6:1024.

    https://doi.org/10.7554/eLife.21231

    • PubMed
    • Google Scholar
19. 1. Iwamatsu T

    (2004) Stages of normal development in the medaka Oryzias latipes

    Mechanisms of Development 121:605–618.

    https://doi.org/10.1016/j.mod.2004.03.012

    • PubMed
    • Google Scholar
20. 1. Jain AK
    2. Barton MC

    (2018) p53: emerging roles in stem cells, development and beyond

    Development 145:dev158360.

    https://doi.org/10.1242/dev.158360

    • PubMed
    • Google Scholar
21. 1. Jonz MG
    2. Fearon IM
    3. Nurse CA

    (2004) Neuroepithelial oxygen chemoreceptors of the zebrafish gill

    The Journal of Physiology 560:737–752.

    https://doi.org/10.1113/jphysiol.2004.069294

    • PubMed
    • Google Scholar
22. 1. Jonz MG
    2. Nurse CA

    (2005) Development of oxygen sensing in the gills of zebrafish

    Journal of Experimental Biology 208:1537–1549.

    https://doi.org/10.1242/jeb.01564

    • PubMed
    • Google Scholar
23. 1. Jungke P
    2. Hammer J
    3. Hans S
    4. Brand M

    (2015) Isolation of novel CreERT2-Driver lines in zebrafish using an unbiased gene trap approach

    PLOS ONE 10:e0129072.

    https://doi.org/10.1371/journal.pone.0129072

    • PubMed
    • Google Scholar
24. 1. Kim D
    2. Langmead B
    3. Salzberg SL

    (2015) HISAT: a fast spliced aligner with low memory requirements

    Nature Methods 12:357–360.

    https://doi.org/10.1038/nmeth.3317

    • PubMed
    • Google Scholar
25. 1. Kizil C
    2. Kyritsis N
    3. Dudczig S
    4. Kroehne V
    5. Freudenreich D
    6. Kaslin J
    7. Brand M

    (2012) Regenerative neurogenesis from neural progenitor cells requires injury-induced expression of Gata3

    Developmental Cell 23:1230–1237.

    https://doi.org/10.1016/j.devcel.2012.10.014

    • PubMed
    • Google Scholar
26. 1. Knopf F
    2. Hammond C
    3. Chekuru A
    4. Kurth T
    5. Hans S
    6. Weber CW
    7. Mahatma G
    8. Fisher S
    9. Brand M
    10. Schulte-Merker S
    11. Weidinger G

    (2011) Bone regenerates via dedifferentiation of osteoblasts in the zebrafish fin

    Developmental Cell 20:713–724.

    https://doi.org/10.1016/j.devcel.2011.04.014

    • PubMed
    • Google Scholar
27. 1. Kragl M
    2. Knapp D
    3. Nacu E
    4. Khattak S
    5. Maden M
    6. Epperlein HH
    7. Tanaka EM

    (2009) Cells keep a memory of their tissue origin during axolotl limb regeneration

    Nature 460:60–65.

    https://doi.org/10.1038/nature08152

    • PubMed
    • Google Scholar
28. 1. Kretzschmar K
    2. Clevers H

    (2016) Organoids: modeling development and the stem cell niche in a dish

    Developmental Cell 38:590–600.

    https://doi.org/10.1016/j.devcel.2016.08.014

    • PubMed
    • Google Scholar
29. 1. Kyritsis N
    2. Kizil C
    3. Zocher S
    4. Kroehne V
    5. Kaslin J
    6. Freudenreich D
    7. Iltzsche A
    8. Brand M

    (2012) Acute inflammation initiates the regenerative response in the adult zebrafish brain

    Science 338:1353–1356.

    https://doi.org/10.1126/science.1228773

    • PubMed
    • Google Scholar
30. 1. Laurent P

    (1984)

    [Morphology and physiology of organs of aquatic respiration in vertebrates: the gill]

    Journal De Physiologie 79:98–112.

    • PubMed
    • Google Scholar
31. 1. Laurent P
    2. Dunel-Erb S
    3. Chevalier C
    4. Lignon J

    (1994) Gill epithelial cells kinetics in a freshwater teleost, Oncorhynchus mykiss during adaptation to ion-poor water and hormonal treatments

    Fish Physiology and Biochemistry 13:353–370.

    https://doi.org/10.1007/BF00003415

    • Google Scholar
32. 1. Leguen I

    (2018) Gills of the medaka (Oryzias latipes): A scanning electron microscopy study

    Journal of Morphology 279:1048.

    https://doi.org/10.1002/jmor.20757

    • Google Scholar
33. 1. Liu J
    2. Tan Y
    3. Zhang H
    4. Zhang Y
    5. Xu P
    6. Chen J
    7. Poh YC
    8. Tang K
    9. Wang N
    10. Huang B

    (2012) Soft fibrin gels promote selection and growth of tumorigenic cells

    Nature Materials 11:734–741.

    https://doi.org/10.1038/nmat3361

    • PubMed
    • Google Scholar
34. 1. Lust K
    2. Sinn R
    3. Pérez Saturnino A
    4. Centanin L
    5. Wittbrodt J

    (2016) De novo neurogenesis by targeted expression of atoh7 to Müller glia cells

    Development 143:1874–1883.

    https://doi.org/10.1242/dev.135905

    • PubMed
    • Google Scholar
35. 1. McKenna A
    2. Findlay GM
    3. Gagnon JA
    4. Horwitz MS
    5. Schier AF
    6. Shendure J

    (2016) Whole-organism lineage tracing by combinatorial and cumulative genome editing

    Science 353:aaf7907.

    https://doi.org/10.1126/science.aaf7907

    • PubMed
    • Google Scholar
36. 1. Mesquita D
    2. Dekanty A
    3. Milán M

    (2010) A dp53-dependent mechanism involved in coordinating tissue growth in Drosophila

    PLOS Biology 8:e1000566.

    https://doi.org/10.1371/journal.pbio.1000566

    • PubMed
    • Google Scholar
37. 1. Meyerowitz EM

    (2002) Plants compared to animals: the broadest comparative study of development

    Science 295:1482–1485.

    https://doi.org/10.1126/science.1066609

    • PubMed
    • Google Scholar
38. 1. Mongera A
    2. Singh AP
    3. Levesque MP
    4. Chen YY
    5. Konstantinidis P
    6. Nüsslein-Volhard C

    (2013) Genetic lineage labeling in zebrafish uncovers novel neural crest contributions to the head, including gill pillar cells

    Development 140:916–925.

    https://doi.org/10.1242/dev.091066

    • PubMed
    • Google Scholar
39. 1. Morgan M

    (1974) Development of secondary lamellae of the gills of the trout, Salmo gairdneri (Richardson)

    Cell and Tissue Research 151:509–523.

    https://doi.org/10.1007/BF00222996

    • PubMed
    • Google Scholar
40. 1. Nassar D
    2. Blanpain C

    (2016) Cancer stem cells: basic concepts and therapeutic implications

    Annual Review of Pathology: Mechanisms of Disease 11:47–76.

    https://doi.org/10.1146/annurev-pathol-012615-044438

    • PubMed
    • Google Scholar
41. 1. Oskarsson T
    2. Acharyya S
    3. Zhang XH
    4. Vanharanta S
    5. Tavazoie SF
    6. Morris PG
    7. Downey RJ
    8. Manova-Todorova K
    9. Brogi E
    10. Massagué J

    (2011) Breast cancer cells produce tenascin C as a metastatic niche component to colonize the lungs

    Nature Medicine 17:867–874.

    https://doi.org/10.1038/nm.2379

    • PubMed
    • Google Scholar
42. 1. Pan YA
    2. Freundlich T
    3. Weissman TA
    4. Schoppik D
    5. Wang XC
    6. Zimmerman S
    7. Ciruna B
    8. Sanes JR
    9. Lichtman JW
    10. Schier AF

    (2013) Zebrabow: multispectral cell labeling for cell tracing and lineage analysis in zebrafish

    Development 140:2835–2846.

    https://doi.org/10.1242/dev.094631

    • Google Scholar
43. 1. Pearson BJ
    2. Sánchez Alvarado A

    (2010) A planarian p53 homolog regulates proliferation and self-renewal in adult stem cell lineages

    Development 137:213–221.

    https://doi.org/10.1242/dev.044297

    • PubMed
    • Google Scholar
44. 1. Quintana E
    2. Shackleton M
    3. Sabel MS
    4. Fullen DR
    5. Johnson TM
    6. Morrison SJ

    (2008) Efficient tumour formation by single human melanoma cells

    Nature 456:593–598.

    https://doi.org/10.1038/nature07567

    • PubMed
    • Google Scholar
45. 1. Reinhardt R
    2. Centanin L
    3. Tavhelidse T
    4. Inoue D
    5. Wittbrodt B
    6. Concordet JP
    7. Martinez-Morales JR
    8. Wittbrodt J

    (2015) Sox2, Tlx, Gli3, and Her9 converge on Rx2 to define retinal stem cells in vivo

    The EMBO Journal 34:1572–1588.

    https://doi.org/10.15252/embj.201490706

    • PubMed
    • Google Scholar
46. 1. Rembold M
    2. Loosli F
    3. Adams RJ
    4. Wittbrodt J

    (2006) Individual cell migration serves as the driving force for optic vesicle evagination

    Science 313:1130–1134.

    https://doi.org/10.1126/science.1127144

    • PubMed
    • Google Scholar
47. 1. Sato T
    2. Vries RG
    3. Snippert HJ
    4. van de Wetering M
    5. Barker N
    6. Stange DE
    7. van Es JH
    8. Abo A
    9. Kujala P
    10. Peters PJ
    11. Clevers H

    (2009) Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche

    Nature 459:262–265.

    https://doi.org/10.1038/nature07935

    • PubMed
    • Google Scholar
48. 1. Schepers AG
    2. Snippert HJ
    3. Stange DE
    4. van den Born M
    5. van Es JH
    6. van de Wetering M
    7. Clevers H

    (2012) Lineage tracing reveals Lgr5+ stem cell activity in mouse intestinal adenomas

    Science 337:730–735.

    https://doi.org/10.1126/science.1224676

    • PubMed
    • Google Scholar
49. 1. Scheres B

    (2007) Stem-cell niches: nursery rhymes across kingdoms

    Nature Reviews Molecular Cell Biology 8:345–354.

    https://doi.org/10.1038/nrm2164

    • PubMed
    • Google Scholar
50. 1. Seery JP

    (2002)

    Stem cells of the oesophageal epithelium

    Journal of Cell Science 115:1783–1789.

    • PubMed
    • Google Scholar
51. 1. Seleit A
    2. Krämer I
    3. Riebesehl BF
    4. Ambrosio EM
    5. Stolper JS
    6. Lischik CQ
    7. Dross N
    8. Centanin L

    (2017) Neural stem cells induce the formation of their physical niche during organogenesis

    eLife 6:e29173.

    https://doi.org/10.7554/eLife.29173

    • PubMed
    • Google Scholar
52. 1. Singh SP
    2. Janjuha S
    3. Hartmann T
    4. Kayisoglu Ö
    5. Konantz J
    6. Birke S
    7. Murawala P
    8. Alfar EA
    9. Murata K
    10. Eugster A
    11. Tsuji N
    12. Morrissey ER
    13. Brand M
    14. Ninov N

    (2017) Different developmental histories of beta-cells generate functional and proliferative heterogeneity during islet growth

    Nature Communications 8:664.

    https://doi.org/10.1038/s41467-017-00461-3

    • PubMed
    • Google Scholar
53. 1. Suetsugu-Maki R
    2. Maki N
    3. Nakamura K
    4. Sumanas S
    5. Zhu J
    6. Del Rio-Tsonis K
    7. Tsonis PA

    (2012) Lens regeneration in axolotl: new evidence of developmental plasticity

    BMC Biology 10:103.

    https://doi.org/10.1186/1741-7007-10-103

    • PubMed
    • Google Scholar
54. 1. Sundin L
    2. Nilsson S

    (2002) Branchial innervation

    Journal of Experimental Zoology 293:232–248.

    https://doi.org/10.1002/jez.10130

    • PubMed
    • Google Scholar
55. 1. Suvà ML
    2. Rheinbay E
    3. Gillespie SM
    4. Patel AP
    5. Wakimoto H
    6. Rabkin SD
    7. Riggi N
    8. Chi AS
    9. Cahill DP
    10. Nahed BV
    11. Curry WT
    12. Martuza RL
    13. Rivera MN
    14. Rossetti N
    15. Kasif S
    16. Beik S
    17. Kadri S
    18. Tirosh I
    19. Wortman I
    20. Shalek AK
    21. Rozenblatt-Rosen O
    22. Regev A
    23. Louis DN
    24. Bernstein BE

    (2014) Reconstructing and reprogramming the tumor-propagating potential of glioblastoma stem-like cells

    Cell 157:580–594.

    https://doi.org/10.1016/j.cell.2014.02.030

    • PubMed
    • Google Scholar
56. 1. Taniguchi Y
    2. Takeda S
    3. Furutani-Seiki M
    4. Kamei Y
    5. Todo T
    6. Sasado T
    7. Deguchi T
    8. Kondoh H
    9. Mudde J
    10. Yamazoe M
    11. Hidaka M
    12. Mitani H
    13. Toyoda A
    14. Sakaki Y
    15. Plasterk RH
    16. Cuppen E

    (2006) Generation of medaka gene knockout models by target-selected mutagenesis

    Genome Biology 7:R116.

    https://doi.org/10.1186/gb-2006-7-12-r116

    • PubMed
    • Google Scholar
57. 1. Trapnell C
    2. Roberts A
    3. Goff L
    4. Pertea G
    5. Kim D
    6. Kelley DR
    7. Pimentel H
    8. Salzberg SL
    9. Rinn JL
    10. Pachter L

    (2012) Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks

    Nature Protocols 7:562–578.

    https://doi.org/10.1038/nprot.2012.016

    • PubMed
    • Google Scholar
58. 1. Tu S
    2. Johnson SL

    (2011) Fate restriction in the growing and regenerating zebrafish fin

    Developmental Cell 20:725–732.

    https://doi.org/10.1016/j.devcel.2011.04.013

    • PubMed
    • Google Scholar
59. 1. Van Keymeulen A
    2. Rocha AS
    3. Ousset M
    4. Beck B
    5. Bouvencourt G
    6. Rock J
    7. Sharma N
    8. Dekoninck S
    9. Blanpain C

    (2011) Distinct stem cells contribute to mammary gland development and maintenance

    Nature 479:189–193.

    https://doi.org/10.1038/nature10573

    • PubMed
    • Google Scholar
60. 1. Vermeulen L
    2. De Sousa E Melo F
    3. van der Heijden M
    4. Cameron K
    5. de Jong JH
    6. Borovski T
    7. Tuynman JB
    8. Todaro M
    9. Merz C
    10. Rodermond H
    11. Sprick MR
    12. Kemper K
    13. Richel DJ
    14. Stassi G
    15. Medema JP

    (2010) Wnt activity defines colon cancer stem cells and is regulated by the microenvironment

    Nature Cell Biology 12:468–476.

    https://doi.org/10.1038/ncb2048

    • PubMed
    • Google Scholar
61. 1. Wilson JM
    2. Laurent P

    (2002) Fish gill morphology: inside out

    Journal of Experimental Zoology 293:192–213.

    https://doi.org/10.1002/jez.10124

    • PubMed
    • Google Scholar
62. 1. Ye X
    2. Tam WL
    3. Shibue T
    4. Kaygusuz Y
    5. Reinhardt F
    6. Ng Eaton E
    7. Weinberg RA

    (2015) Distinct EMT programs control normal mammary stem cells and tumour-initiating cells

    Nature 525:256–260.

    https://doi.org/10.1038/nature14897

    • PubMed
    • Google Scholar

Thank you for submitting your article "Hierarchical stem cell topography splits growth and homeostatic functions in the fish gill" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by Alejandro Sánchez Alvarado as the Reviewing Editor and Didier Stainier as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

This carefully executed and richly described effort provides evidence for the existence of a complex stem cell population required for the maintenance and growth of gills in the Japanese rice fish Medaka. The authors have identified a very interesting system in which to study stem cell population dynamics in an adult organ during both injury and homeostasis. By using a combination of lineage tracing and pulse-chase experiments under homeostatic and injury conditions, the authors identified at least 4 unipotent and developmentally restricted adult stem cell types. The documentation of the distinct stem cell systems, in this almost two-dimensional cellular array, is an important finding. In sum, this is a simple and elegant stem cell system that has great potential to uncover novel mechanisms of how one organ system balances growth and regeneration.

Essential revisions:

While all of the reviewers are in agreement of the accessibility and potential of the gill system to inform and uncover dynamics of stem cell regulation during homeostasis and regeneration, there is also agreement that the manuscript in its present form is devoid of molecular detail. As such, the reviewers would appreciate the incorporation of any molecular data the authors may already have that could help address the comments below.

1) The authors describe when the fate of homeostatic and growth aSCs are fixed and locate in the tissue, but they do not show any molecular mechanism of how this is achieved. in situ, qpcr, profiling studies to identify specific stem cells markers in different zone and time points are missing. Could not label retention be used here to identify and isolate these cells and characterize them in some detail?

2) The idea of modifying of the microenvironment is excellent, but it is not clear how specific and accurate the ablation method is. What about the surrounding cells? Controls were needed here.

3) Figures 2 and 4 appear to provide different conclusions. In Figure 2 lineage tracing – the labeled stem cell can give rise to all cells of the filaments. In figure 4, the authors show that in their lineage tracing experiments – 4 distinct unipotent stem cells can give rise to the different patterns observed.

a) Is this difference due to labeling strategy?

b) Why was day 9 chosen for tamoxifen?

c) Does the tip stem cell and other unipotent stem cells arise from a single cell that becomes developmentally restricted?

4) In Figure 4 – the authors suggest that the gill has 4 unipotent stem cells. Evidence supporting that the labeling strategy employed is truly only labeling one cell in their model system is required. This is an important point as these experiments will help determine whether a cell type within the filaments arises from a single labeled cell. Otherwise, how would these specific patterns arise?

5) In Figure 7, the authors ablate filaments and then lineage-trace to determine whether regeneration happens. In the lineage traced animals, the authors identified 4 patterns. Did the authors ablate filaments from all types of lineage-traced patterns? Can all unipotent type of adult stem cells regenerate the tip? Or was the regeneration only seen from pattern 1 animals? Also, is it possible to distinguish between unipotent stem cells driving regeneration or if the differentiated cells re-enter the cell cycle? Are there markers for these cells that can be used for co-staining?

Essential revisions:

While all of the reviewers are in agreement of the accessibility and potential of the gill system to inform and uncover dynamics of stem cell regulation during homeostasis and regeneration, there is also agreement that the manuscript in its present form is devoid of molecular detail. As such, the reviewers would appreciate the incorporation of any molecular data the authors may already have that could help address the comments below.

We are glad to hear that the reviewers appreciate the potential of the gill system, and we understand their concerns about missing molecular data to complement the extensive functional analysis on gill stem cells that we presented in our initial submission. We have improved the revised version by adding two datasets that provide molecular information at different scales.

The first addition complements our functional characterisation of growth stem cell at the tip of the filament. We have conducted RNA-seq on two different regions along gill filaments: the filament trunk, enriched in differentiated cell, and the region that we identified as the filament growth domain, enriched in undifferentiated cells. The analysis of the transcriptomes has allowed us to validate molecularly what we had described functionally. We found numerous genes involved with the physiological function of the gill differentially expressed in the filament trunk, while differentially expressed genes at the growth domain were identified as “non-differentiated tissue”. The analysis has been carried out by two colleagues (L.B. and J-R.M.M.), who are included in the author list of the revised manuscript. The data can be found at the new Supplementary file 5, new Figure 3G, new Figure 3—source data 1, and in the new last paragraph of the section “Growth Stem Cells Locate to the Growing Edge of Each Filament”. The raw data has been deposited at: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE130939.

The second addition tackles the functional role of the p53 gene during the coordinated post-embryonic growth of the gill filaments. We have used p53/WT chimeras to generate composite filaments, which allowed us to reveal a lineage specific function of P53. Briefly, filaments composed of p53 mutant Pattern 1, 2 and 3 and a WT Pattern 2 display a growth defect when compared to their immediate neighbour filaments with other composition of p53 and WT Patterns. These experiments position p53 as an important regulator of growth stem cells in the fish gill, and reveal a differential contribution of the different lineages to overall growth of the filament. We have included these results in the new Figure 6, new Figure 6—figure supplement 1, and new Section “Wild type – p53-/- chimeras reveal functional differences among lineages”.

We understand that these additional analyses largely complement the main message of our manuscript and want to thank the reviewers for the constructive suggestions.

1) The authors describe when the fate of homeostatic and growth aSCs are fixed and locate in the tissue, but they do not show any molecular mechanism of how this is achieved. in situ, qpcr, profiling studies to identify specific stem cells markers in different zone and time points are missing. Could not label retention be used here to identify and isolate these cells and characterize them in some detail?

We have followed the suggestion of the reviewers to better characterise molecularly gill stem cells, and include in our new version an RNA-seq analysis of different domains along the filaments (new Supplementary file 5, new Figure 3G, and in the new last paragraph of the section “Growth Stem Cells Locate to the Growing Edge of Each Filament”, see previous response).

We have also tried label retention experiments with BrdU (which were indeed used in the initial submission, Figure 3A, E, F), but these were not useful to detect label retention cells in the growth domain. We believe that this is due to the fact that the filaments contain a very small number of stem cells for each lineage, and their high proliferative activity dilutes any label that stem cells can acquire during the BrdU treatment. We are indeed following a bioinformatic approach to estimate the number of stem cells per filament and per brachial arch (with D-PD and AMC), based on the proportion of cells labelled per lineage when a filament displays a labelled pattern. Preliminary analysis suggest that Patterns 2, 3 and 4 are maintained by a single stem cell (since we always get an all-or-none labelling after recombination approached in filament stem cells), and Pattern 2 must contain between 2 and 3.

The low number of growth stem cells per filament also impacts on our RNA-seq analysis on bulk samples. We have detected an expression profile that matches an undifferentiated domain (new Supplementary file 5 and new last paragraph of the section “Growth Stem Cells Locate to the Growing Edge of Each Filament”), but it is clear that a bulk analysis on such sample will most likely reflect the transcriptional profile of amplifying progenitors rather than the one of bona-fide stem cells. Further analysis of single-cell RNA-seq will certainly improve our understanding on molecular pathways controlling growth in each lineage and growth coordination among lineages, and provide a narrower number of candidate genes to test functionally. We believe the work we present here, revealing three different levels of stem cells and 4 types of fate-restricted lineages for each level, constitutes a necessary platform for future, detailed molecular analysis of individual lineages in a constantly growing composite organ. The new, exciting results that we provide using existing p53 mutants (new Figure 6, new Figure 6—figure supplement 1, and new Section “Wild type – p53-/- chimeras reveal functional differences among lineages”) prove the potential of the gill system to understand cross-lineage communication and identify a main factor involved in this coordination.

2) The idea of modifying of the microenvironment is excellent, but it is not clear how specific and accurate the ablation method is. What about the surrounding cells? Controls were needed here.

We understand that the ablation experiments are ill-described in the current version of our manuscript and added the relevant additional information for the revised version, which we believe helps the reader understanding the system better. Filaments are ablated using small surgery scissors on anaesthetised fish, accessing the gill by opening the operculum with dedicated forceps. Filaments are densely packed along a branchial arch, but they conform detached units all along their longitudinal axis. This means that filaments can be treated independently, and the ablation of one of them does not affect the microenvironment on the growth region of neighbour filaments.

We used for this experiment internal controls, i.e. neighbour non-ablated filaments along the same branchial arch, and also the corresponding filaments from the contralateral branchial arch. Both types of control filaments keep growing at the pace of filaments in a control gill, which can be assessed by comparing their relative size along the branchial arch.

To give more visibility to this aspect, we have included new sentences in the Discussion, where we stress the fact that filament growth is an autonomous – i.e., a filament growth rate does not adjust to the size of its neighbour filaments. We believe that this observation is particularly intriguing considering the order in filament size along a branchial arch, where every filament is larger that its younger neighbour and shorter than its older neighbour. Our observations in the regeneration section, and the new results with p53-WT chimeras presented in the revised version, indicate that this stunning distribution is achieved by the autonomous growth of each unit, which positions branchial arches as ideal systems to explore growth coordination – a line of research that we are following and hoping to communicate in the near future.

3) Figures 2 and 4 appear to provide different conclusions. In Figure 2 lineage tracing – the labeled stem cell can give rise to all cells of the filaments. In Figure 4, the authors show that in their lineage tracing experiments – 4 distinct unipotent stem cells can give rise to the different patterns observed.

a) Is this difference due to labeling strategy?

b) Why was day 9 chosen for tamoxifen?

c) Does the tip stem cell and other unipotent stem cells arise from a single cell that becomes developmentally restricted?

We appreciate the comment from the reviewers to help us clarify this important aspect of the post-embryonic growth of the gills. Our results are consistent all along the manuscript and show that filament growth is driven by 4 different fate-restricted stem cells.

The panels in Figure 2 correspond to an entire gill (2B) and en entire branchial arch (2C) that were imaged using a fluorescent binocular, and therefore do not allow characterising clones at the cellular resolution. Figure 2C show filaments labelled in the Pattern1, which is the one comprising the most abundant number of cells, and can be missed for a fully labelled filament. Figure 2B shows filaments labelled in different patterns, but here again the patterns containing more labelled cells (Pattern1 and Pattern2) are the most prominent, easier to detect. The data in Figures 3D, 4, 5, 6, 7, 8 and Figure 5—figure supplement 1 was acquired using confocal microscopes, and they all show lineage restriction. We therefore added a sentence clearly stating this fact in the main text (at the time we describe the results in Figure 2) and in the figure legends to avoid confusions during the initial characterisation of stem cell domains.

In addition, and to answer the points a and c, we include in the revised version additional data using alternative methods of labelling that result in the very same scenario: at least 4 different types of stem cells maintaining growth or each filament and each branchial arch during medaka post-embryonic life. The results presented in the initial version were obtained by inducing recombination on Gaudí^Ubiq.iCRE Gaudí^RSG transgenic fish with tamoxifen at day 9 pf. This developmental stage was chosen as it is the youngest stage of post-embryonic life.

We added now earlier inductions indicating that gills are formed by stem cells that are already fate-restricted before the organ is formed, going against the presence of an initial common stem cell for all lineages in the fish gill. We demonstrate this by using Gaudí^Hsp70A.CRE Gaudí^RSG transgenic fish to induce recombination at different embryonic stages (from stage 20, ca. 2 dpf., to stage 34, ca. 4 dfp). We have also performed transplantations at the blastula stage, which are arguably the earliest way to generate proper lineage clones in fish, using the truly ubiquitous H2B-EGFP transgenic line Gaudí^LoxPOUT. All these experiments resulted in the same scenario than the previously described, showing the very same recombination patterns in the gill filaments. We have found cases in which all filaments in a branchial arch are labelled in the same pattern (new Figure 5—figure supplement 1B)(Pattern 1 N=1 branchial arch, Pattern 3 N=2 branchial arches, Pattern 4 N=1 branchial arch), indicating that the earliest cell for a given lineage is not contributing to any of the other lineages and these therefore constitute different units. We included this new data in the new Figure 5—figure supplement 1, and in the new last paragraph of the section “Growth Stem Cells are Fate Restricted”.

4) In Figure 4 – the authors suggest that the gill has 4 unipotent stem cells. Evidence supporting that the labeling strategy employed is truly only labeling one cell in their model system is required. This is an important point as these experiments will help determine whether a cell type within the filaments arises from a single labeled cell. Otherwise, how would these specific patterns arise?

This is an important aspect, which we have already partially answered in our previous responses (Answer 1 and Answer 3). In brief, we can not target recombination to one specific filament stem cell in the fish gill. We have, however, tuned tamoxifen concentrations down, reduced the temperature difference during heat-shock inductions, and transplanted fewer cells in our different clone-generating strategies. And in all these samples, we have focused on recombined gills that showed reduced recombination events (this is clearly stated in the main text and the new Materials and methods section of the revised version). We have also included mathematical models to assess the minimal number of cells that are compatible with our experimental observations, and these revealed that the system uses indeed a few number of stem cells per lineage. This is carried out by computing the likelihood of independent recombination events that would result in a stretch of 6 consecutive filaments having the same recombination pattern. We have analysed dozens of gills containing hundreds of labelled filaments and the data obtained adjusts to models that use between 1 and 3 active stem cells per lineage in a given branchial arch (Data presented in Figure 2D and Supplementary files 1-4, 6). In addition, we have used LoxP reporter lines that have additional colours as read outs for recombination (the so-called Brainbow / Confetti principle) and we have observed the same lineage-restriction reported using monochromatic LoxP read-outs. Since these lines contain fluorescent proteins that are not localised to the nucleus, matching these samples to the data we show using H2B-EGFP is non-trivial and that is why we do not include those in the present manuscript.

5) In Figure 7, the authors ablate filaments and then lineage-trace to determine whether regeneration happens. In the lineage traced animals, the authors identified 4 patterns. Did the authors ablate filaments from all types of lineage-traced patterns? Can all unipotent type of adult stem cells regenerate the tip? Or was the regeneration only seen from pattern 1 animals? Also, is it possible to distinguish between unipotent stem cells driving regeneration or if the differentiated cells re-enter the cell cycle? Are there markers for these cells that can be used for co-staining?

We have indeed ablated filaments that contained labelled cells of Pattern 1, Pattern 2 and Pattern 3 (results presented in Figure 8G-I of the revised version). All of them showed the same trend, where recombination patterns are maintained in the new domain formed during the regeneration phase and indicating that the stem cells generating Patters 1-to-3 do not change their fate upon injuries. As previously mentioned, recombination Pattern 4 is difficult to be observed in living samples and this is why we could not perform target ablations on these filaments.

The questions on whether regeneration is triggered by homeostatic stem cells that change to a growth mode, or alternatively, by differentiated cells that re-enter the cell cycle is fascinating and highly relevant in the regeneration field, as both cases have been reported. The ultimate experiment to tackle this question implies using cell-type specific promoters to trigger recombination in post-mitotic cells of the different lineages to then assay whether the regenerative part of a filament derived from labelled or non-labelled cells. We do not have specific markers so far and therefore this particular experiment could not be performed for the revision. We are confident that the use of single-cell RNA-seq will give us valid candidates to approach this aspect for the fish gill. What we can be sure of, is that the regeneration of each individual lineage is dealt with cells of the same lineage. This point is mentioned in the Discussion of the revised version.

Author details

1. Julian Stolper

   1. Centre for Organismal Studies, Heidelberg University, Heidelberg, Germany
   2. Murdoch Children’s Research Institute, Royal Children’s Hospital, Parkville, Australia

   Contribution

   Data curation, Formal analysis, Visualization, Writing—review and editing

   Competing interests

   No competing interests declared

2. Elizabeth Mayela Ambrosio

   Centre for Organismal Studies, Heidelberg University, Heidelberg, Germany

   Present address

   The Novo Nordisk Foundation Center for Stem Cell Research, University of Copenhagen, Copenhagen, Denmark

   Contribution

   Data curation, Formal analysis, Visualization, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7227-7744

3. Diana-Patricia Danciu

   Institute of Applied Mathematics, Heidelberg University, Heidelberg, Germany

   Contribution

   Formal analysis, Methodology, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8683-3956

4. Lorena Buono

   Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide, Seville, Spain

   Contribution

   Resources, Data curation, Formal analysis, Visualization, Methodology

   Competing interests

   No competing interests declared

5. David A Elliott

   Murdoch Children’s Research Institute, Royal Children’s Hospital, Parkville, Australia

   Contribution

   Resources, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-1052-7407

6. Kiyoshi Naruse

   Laboratory of Bioresources, National Institute for Basic Biology, National Institutes of Natural Sciences, Okazaki, Japan

   Contribution

   Resources, Data curation

   Competing interests

   No competing interests declared

7. Juan R Martínez-Morales

   Centro Andaluz de Biología del Desarrollo, Universidad Pablo de Olavide, Seville, Spain

   Contribution

   Resources, Data curation, Formal analysis, Visualization, Methodology

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4650-4293

8. Anna Marciniak-Czochra

   1. Institute of Applied Mathematics, Heidelberg University, Heidelberg, Germany
   2. Bioquant Center, Heidelberg University, Heidelberg, Germany

   Contribution

   Data curation, Formal analysis, Supervision, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5831-6505

9. Lazaro Centanin

   Centre for Organismal Studies, Heidelberg University, Heidelberg, Germany

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Visualization, Writing—original draft

   For correspondence

   lazaro.centanin@cos.uni-heidelberg.de

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-3889-4524

• 2,580

  Page views

• 221

  Downloads

• 5

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.