To begin to elucidate the mechanisms of RNE formation, we initially focused on RNE invagination. It has been recently shown that RNE invagination is accompanied by basal constriction of neuroepithelial cells in the distal layer of the optic vesicle (Figure 1A, brown cell) (Nicolás-Pérez et al., 2016). To validate this finding, we labeled the cell cortex at different stages of optic cup morphogenesis from 15 somite stage (ss) to 28 ss (16 hours post fertilization (hpf) to 24 hpf) using the F-actin marker phalloidin. We measured the average cell density at the base of the RNE and noted that basal cell density increased significantly during optic cup invagination (Figure 1—figure supplement 1A,C). Concomitantly, the average basal area of RNE cells was reduced by about 45%. This reduction mainly occurred during the first half of the invagination process between 15 ss and 21 ss (Figure 1B). In contrast, the average apical area of cells increased. As a result, the average apical area was larger than the average basal area at the end of the process. This shows that cells undergo overall shape changes during invagination and form cones with a narrowing basal surface (Figure 1B).

To determine whether the observed reduction in basal cell area was linked to changes in actomyosin distribution, we imaged actin and myosin distribution during RNE morphogenesis. At the onset of neuroepithelial invagination, the actin marker GFP-UtrophinCH (Burkel et al., 2007) accumulated at the basal side of the RNE (Video 1, Figure 1C) similarly to the myosin marker, myl12.1-EGFP (Iwasaki et al., 2001) (Video 2, Figure 1D). Basal actin and myosin enrichment was specific to the invaginating zone of the neuroepithelium and was not observed in other regions of the optic cup (Videos 1 and 2, Figure 1C,D). These findings were corroborated by phalloidin (actin) and anti-phosphomyosin (active myosin) staining of samples fixed between 15 ss and 30 ss (Figure 1E,G and Figure 1—figure supplement 1B). Quantitative analysis of phalloidin and phosphomyosin intensity distribution along the apicobasal axis of the neuroepithelium showed that the basal actomyosin bias spanned about 15% of the RNE height (Figure 1F,H and Figure 1—figure supplement 1D). Thus, we confirmed that invagination of the neuroepithelium is accompanied by basal surface area reduction (Nicolás-Pérez et al., 2016). We further show that invaginating RNE cells change their overall shape and feature basal accumulation of contractile actomyosin (Figure 1F and H).

Video 1

Download asset

Video 2

Download asset

To test whether the pool of contractile basal actomyosin is actively involved in RNE invagination, we perturbed myosin contractility using the drug Rockout, which is known to inhibit the Rock pathway upstream of actomyosin activity (Yarrow et al., 2005). Upon 7 h of Rockout treatment, lens structures in treated embryos developed similarly to controls (Figure 2A). In contrast, the basal actomyosin bias in invaginating RNE cells was markedly reduced (Figure 2A,C,D). Also, the overall staining for phosphomyosin was weaker in the treated embryos (Figure 2—figure supplement 1A). Despite this reduction of actin and myosin however, the effect on the morphology of the cup was surprisingly weak. 40% of embryos showed an accumulation of cells at the rim of the cup, whereas 30% of embryos showed a milder optic cup phenotype (n = 30 embryos per experiment, N = 5 experiments; 30% of embryos showed stalled development and were thus excluded from the analysis; Figure 2A,B, see Figure 2—source data 1). The invagination angle (defined as the angle between the inner lips of the optic cup and the center of the RNE, see schematic in Figure 2F and Material and methods) was only moderately wider in Rockout treatement with an average of 80° (SEM ± 3.979 n = 15) compared to 60° (SEM ± 2.631 n = 14) in DMSO controls (Figure 2F). Thus, RNE invagination was initiated but was not as efficient in the treated embryos as in controls and appeared delayed (Figure 2A,F). To better understand the progression of RNE formation in the Rockout condition, we performed time-lapse imaging of GFP-UtrophinCH in Rockout-treated embryos. This confirmed that RNE invagination still took place, albeit with slower kinetics, even when basal actin levels were greatly reduced (Video 3, Figure 2E,G). Interestingly, at 30ss, the end point of optic cup invagination, the basal surface area of the invaginating cells was similar in controls and Rockout-treated embryos (Figure 2G). This observation suggests that RNE invagination does not solely depend on basal actomyosin contractility and that it can be compensated by other mechanisms.

Figure 2 with 1 supplement see all

Download asset Open asset

Figure 2—source data 1

Related to Figure 2A, referred in text.

https://doi.org/10.7554/eLife.22689.010

Download elife-22689-fig2-data1-v2.xlsx

Figure 2—source data 2

Related to Figure 2C.

https://doi.org/10.7554/eLife.22689.011

Download elife-22689-fig2-data2-v2.xlsx

Figure 2—source data 3

Related to Figure 2D.

https://doi.org/10.7554/eLife.22689.012

Download elife-22689-fig2-data3-v2.xlsx

Figure 2—source data 4

Related to Figure 2F.

https://doi.org/10.7554/eLife.22689.013

Download elife-22689-fig2-data4-v2.xlsx

Figure 2—source data 5

Related to Figure 2G.

https://doi.org/10.7554/eLife.22689.014

Download elife-22689-fig2-data5-v2.xlsx

Video 3

Download asset

RNE invagination could also be aided by cell compaction when an increasing number of cells inhabit the RNE and these cells contribute to bending of the basal surface. One way to increase cell numbers is proliferation, and it has been shown that RNE formation is accompanied by high proliferation rates (Kwan et al., 2012). In addition, proliferation has been suggested as a driving force for optic cup formation in vitro (Eiraku et al., 2011). In zebrafish and Xenopus, however, optic cup invagination persists even when proliferation is inhibited (Kwan et al., 2012; Harris and Hartenstein, 1991). To test whether tissue compaction caused by ongoing proliferation can at all influence RNE invagination, we inhibited proliferation with hydroxyurea and aphidicolin from 10 ss onwards (Kwan et al., 2012), leading to markedly reduced pH3 staining compared to controls (Figure 2—figure supplement 1B). Consequently, the neuroepithelium hosted fewer cells and these cells adopted a more columnar morphology than control cells (Figure 2—figure supplement 1C,D and [Kwan et al., 2012]). Furthermore, following inhibition of proliferation, cells in the RNE displayed a larger average basal area than control cells (Figure 2G). Despite this, optic cup formation still occurred and invagination angles were not severely affected (Figure 2F and Figure 2—figure supplement 1C). This observation indicated two things a) That the total number of cells in the invaginating RNE influences the basal area of single cells and b) that RNE invagination can take place even when cells show a broader basal area due to reduced cell proliferation. Thus, it suggests that other additional processes participate in RNE morphogenesis.

In addition to proliferation, another process that occurs concomitantly with RNE invagination that increases the cell number in the RNE is rim cell involution. This phenomenon is most prominent at the ventral and temporal side of the cup and relocates a substantial number of cells from the proximal layer of the optic vesicle into the invaginating RNE (blue cell, Figure 1A) (Kwan et al., 2012; Picker et al., 2009; Heermann et al., 2015). Such an influx of cells may result in increased tissue compaction and contribute to the inward bending of the tissue. As we noted that rim cells in the Rockout-treated embryos continued to move into the cup (Figure 2E, Video 3), we wondered whether these cell movements could explain the recovery of RNE morphogenesis defects upon Rockout treatment. We thus asked if rim involution was actively involved in shaping the RNE. To answer this question, we first needed to understand how cells move at the rim. Hence, we characterized their single-cell dynamics. Mosaic labeling of rim cells with the membrane marker mKate2-ras and the apical marker Par3-GFP revealed that moving rim cells kept a stable apical domain but were very dynamic at the basal side (Figure 3A, Video 4). As microtubules marked with EMTB-tdTomato did not show significant reorganization during rim involution, we concluded that microtubules were not majorly involved in the cellular dynamics (Figure 3—figure supplement 1A). Instead, we observed that cells extended dynamic lamellipodia-like protrusions in the direction of movement (Figure 3A, Video 4). These protrusions were actin rich, as shown by GFP-UtrophinCH localization (Figure 3B, Video 5), underlining their lamellipodial character. Such lamellipodia were only observed in the rim cells and were not seen in the invaginating RNE cells (Figure 3B, Video 5). When we quantified basal protrusive activity during rim migration, we found that the vast majority of the actin rich protrusions occurred in the direction of movement, the leading side of the cells facing the invaginating zone (Figure 3D,E). This showed that the protrusive activity of rim cells is directional. Interestingly, the directional actin-rich lamellipodia formation and rim migration were still observed in the Rockout condition (Figure 3F,G, Video 6), further indicating that the migratory behavior of rim cells occurs independently of the basal actomyosin bias in the invaginating RNE cells. Interestingly, while rim cells exhibited protrusive membrane activity at the basal side, they nevertheless maintained their apicobasal polarity and an apical actin pool associated with the adherens junctions (Figure 3C, Videos 7 and 8). This shows that a) rim cells are integrated in an intact epithelial sheet and display the migratory behavior as a collective epithelium and b) that rim cells show characteristics that are intermediate between the epithelial and mesenchymal state.

Figure 3 with 1 supplement see all

Download asset Open asset

Figure 3—source data 1

Related to Figure 3E,F.

https://doi.org/10.7554/eLife.22689.018

Download elife-22689-fig3-data1-v2.xlsx

Video 4

Download asset

Video 5

Download asset

Video 6

Download asset

Video 7

Download asset

Video 8

Download asset

However, once rim cells reached the RNE, basal actin dynamics changed considerably and mesenchymal characteristics were lost. Lamellipodia formation ceased and a pool of stable basal actin was observed similar to the actin distribution in the invaginating RNE cells (Figure 3B, Video 5). Likewise, while myosin labeled by DD-myl12b-GFP occurred as dynamic spots at the basal side of cells during rim movement (Video 9, Figure 3—figure supplement 1B), it became stably enriched once the cells reached the inner side of the optic cup. Together, these data show that rim cells exhibit dynamic and directional basal lamellipodia formation that coincides with directed migratory behavior. Once cells reach the RNE, however, they change their basal dynamics and establish a stable basal actomyosin pool similar to the basally constricting RNE cells.

Video 9

Download asset

Our analysis so far suggested that while being integrated in an epithelium, rim cells exhibit migratory behavior as a collective driven by directed basal lamellipodial activity. We next asked how exactly lamellipodia formation can drive rim cell movement. Lamellipodia-driven migration often depends on dynamic attachments of cells to the surrounding extracellular matrix (ECM) (Friedl and Wolf, 2010). Thus, we assessed the distribution of ECM during rim movement by immunostainings for the ECM components Laminin, Chondroitin sulfate and Fibronectin. Laminin and Chondroitin sulfate were only weakly expressed beneath the developing retinal pigment epithelium (RPE), but showed stronger accumulation in the rim zone, the developing lens placode and the invaginating RNE (Figure 4—figure supplement 1A). Fibronectin instead showed preferential staining in the rim zone and the developing lens placode and weaker staining beneath the invaginating RNE (Figure 4—figure supplement 1B). Hence, we speculated that the migrating rim cells use their lamellipodia to dynamically attach to the underlying ECM and thereby generate force for movement. Imaging the focal adhesion markers Integrin-β1-mKate2 and Paxillin-mKate2 indeed revealed dynamic short-lived foci at the basal side of rim cells during time-lapse imaging (Videos 10 and 11, Figure 4A,B). However, once cells reach the RNE and stopped the active migratory behavior they formed stable focal adhesions marked by integrin and paxillin as well as actin (Videos 10 and 11, Figure 4A,B). Co-imaging of the localization and dynamics of Paxillin-mKate2 and GFP-UtrophinCH corroborated this finding. While the cells in the rim zone lacked basal enrichment of actin and paxillin, cells in the invaginating region showed stable basal enrichment of Paxillin coinciding with the basal accumulation of actin (Figure 4—figure supplement 1C, Video 12). This suggested that the active migratory behavior observed for rim cells is confined to the zone that connects the developing RPE and the RNE, where cells move from one epithelial layer into the other by turning 180°. Once the cells reach the inner layer of the cup they change to stable cell-ECM attachment and attain typical neuroepithelial morphology.

Video 10

Download asset

Video 11

Download asset

Figure 4 with 3 supplements see all

Download asset Open asset

Figure 4—source data 1

Related to Figure 4—figure supplement 2D.

https://doi.org/10.7554/eLife.22689.029

Download elife-22689-fig4-data1-v2.xlsx

Figure 4—source data 2

Related to Figure 4G.

https://doi.org/10.7554/eLife.22689.030

Download elife-22689-fig4-data2-v2.xlsx

Figure 4—source data 3

Related to Figure 4—figure supplement 3C.

https://doi.org/10.7554/eLife.22689.031

Download elife-22689-fig4-data3-v2.xlsx

Figure 4—source data 4

Related to Figure 4—figure supplement 3D.

https://doi.org/10.7554/eLife.22689.032

Download elife-22689-fig4-data4-v2.xlsx

Video 12

Download asset

To verify that such dynamic interactions with the ECM were indeed necessary for successful rim cell migration, we perturbed cell-matrix attachment by interfering with the ECM or the intracellular attachment side. To this end, we knocked down laminin α-1 and ojoplano/opo using previously published morpholinos (Martinez-Morales et al., 2009; Pollard et al., 2006) (see Figure 4—figure supplement 2A–D for controls for efficiency of morpholino knockdown). Laminin depletion interferes directly with ECM integrity and it has been shown to be involved at different stages of optic cup development (Ivanovitch et al., 2013; Bryan et al., 2016). We observed that depletion of laminin from the ECM resulted in an overall increase in Fibronectin staining, indicating a change in the ECM composition and arrangement (Figure 4—figure supplement 3A). Knockdown of laminin resulted in impaired optic cup formation at 30 ss. Epithelial tissue accumulated at the temporal side of the optic cup was observed, giving the cup an S-shaped appearance instead of the C-shaped appearance seen in controls (Figure 4C,D).

To test if intracellular perturbation of cell-ECM attachment would also affect rim migration, we knocked down Opo. Opo is a protein that is known to regulate endocytosis of integrins and was previously shown to play an important role in eye morphogenesis (Martinez-Morales et al., 2009; Bogdanović et al., 2012). Indeed, opo morphants recapitulated the S-shaped optic cup phenotype, although the severity of the phenotype was milder than in the laminin condition (Figure 4C,D).

Time-lapse imaging of both conditions confirmed that the S-shaped phenotype was caused by impaired rim cell migration, as the speed of rim cells was reduced (Videos 13 and 14, Figure 4E,F,G). This reduced speed most likely results from the inability of migrating cells to efficiently use their lamellipodia to migrate along the underlying matrix. A recent study reported that Laminin α-1 mutant shows polarity defects in cells of the optic cup (Bryan et al., 2016). However, in the morphant condition used here we did not observe a strong effect on polarity. We only observed few ectopic apical domains within the neuroepithelium while overall tissue-wide apical marker distribution was not affected in the laminin morphants (Figure 4—figure supplement 3B). Thus, this condition allowed us to dissect the effects of perturbed cell-ECM adhesion during OCM. Similarly, no defects on apical marker distribution were observed in Opo morphants (Figure 4—figure supplement 3B, Martinez-Morales et al., 2009). Basally however, we noticed that actomyosin accumulation was reduced in the invagination zone in both laminin and opo knockdown conditions (Figure 4D, Figure 4—figure supplement 3C,D). In addition, invagination angles were affected (Figure 2F). This argued that both, rim migration and actomyosin-assisted invagination were disturbed in these conditions leading to the strong phenotypes observed.

Video 13

Download asset

Video 14

Download asset

Together, these data imply that interference with cell-ECM attachment leads to disturbance of the two cell behaviors involved in hemispheric RNE formation, rim cells cannot actively migrate due to impaired cell ECM-attachments and otherwise invaginating RNE cells failed to establish the basal actomyosin bias.

To elucidate the cellular changes that led to the impaired rim cell migration upon interference with cell-ECM attachment, we analyzed the basal membrane dynamics of rim cells in the laminin and opo knockdown conditions. When we imaged the focal adhesion component Paxillin using mosaic Paxillin-mKate2 RNA injection, we observed that in laminin morphants the rim cells localized Paxillin stably along the modified basement membrane instead of the dynamic foci seen in the control condition (Video 15, Figure 5—figure supplement 1A). In opo morphants however, Paxillin distribution was more transient than seen in the controls (Video 16, Figure 5—figure supplement 1B). In both cases the rim cells exhibited basal bleb-like protrusions instead of the lamellipodia observed in control embryos (Videos 15 and 16, Figure 5—figure supplement 1A,B). These bleb-like protrusions were extension of the plasma membrane, with short lived actin localization and devoid of myosin, showing the characteristic features of blebs (Figure 5—figure supplement 1C). Such basal blebs were also observed when opo morphants cells were transplanted into control embryos, in line with a cell autonomous phenotype (Video 17, Figure 5—figure supplement 1D, N = 4 out of 5 transplanted embryos). The observation that morphant cells nevertheless reached the RNE underlines the collective characteristic of the migration as surrounding control cells most likely pull them along to their final positions (Video 17). The finding that the cells that failed to undergo active rim migration exhibited blebs instead of lamellipodia made us ask whether perturbation of the normal protrusive activity itself could interfere with rim cell migration. To test this idea, we knocked down the ERM protein family member Ezrin using a published morpholino (Link et al., 2006). Ezrin is a protein that links actin and the plasma membrane and its knockdown has been shown to lead to increased blebbing and reduced migration rates of zebrafish prechordal plate precursor cells (Schepis et al., 2012) (see Figure 5—figure supplement 2A,B for controls for efficiency of morpholino knockdown by Western blot and overall morphant morphology). During RNE formation, we observed that ezrin knockdown did not affect the overall apicobasal polarity in the developing optic cup (Figure 5—figure supplement 2C) but specifically induced basal blebbing in the area where rim cells resided, whereas lateral sides of the rim cells and the pool of invaginating RNE cells were not affected (Video 18, Figure 5A). We also observed rim cell blebbing when ezrin morphant cells were transplanted into control background, again arguing that blebbing is a cell autonomous phenotype ([Figure 5—figure supplement 2D] N = 2 out of 4 transplanted embryos. This variability is most likely due to varying morpholino concentration within transplanted cells as they come from different donors). Interestingly, ezrin depletion in all rim cells resulted in a phenotype similar to Laminin or Integrin interference. The average cell speed was reduced by 27% and rim cells did not reach the RNE (Figures 4G and 5B,C and F, Video 19). While the formation of the lens and the RPE occurred similar to the control situation (Figure 5C), specifically RNE morphogenesis in ezrin morphants was defective and at 30 ss an S-shaped optic cup was observed (Figure 5B,C). Importantly, however, we only observed a minor effect on basal actomyosin accumulation in the RNE (Figure 5D,E). A comparison of the instantaneous slopes of the normalized phalloidin intensity distribution curves confirmed that unlike Rockout, laminin Mo and Opo Mo, ezrin morphants showed very little disturbance of the basal actomyosin (Figure 5—figure supplement 2E). This indicated that the S-shaped phenotype was mainly caused by the defects in rim involution and less by the perturbation of basal actomyosin. This S-shaped phenotype did not recover upon continued development (Figure 5B). In contrast, the cells that stayed outside the presumptive RNE layer eventually showed a stable Paxillin distribution outside the RNE as seen in the laminin knock down condition (Figure 5G,H). Cells thereby attained neuroepithelial-like morphology. We conclude that correct ECM formation and distribution is important for lamellipodia formation and successful rim cell migration. When cell-ECM attachment is impaired, lamellipodia formation is perturbed and rim cells form blebs but these blebs cannot support cellular movement. Consequently, cells do not enter the neuroepithelium but can nevertheless take on neuroepithelial morphology at ectopic location.

Video 15

Download asset

Video 16

Download asset

Video 17

Download asset

Video 18

Download asset

Figure 5 with 2 supplements see all

Download asset Open asset

Figure 5—source data 1

Related to Figure 5E.

https://doi.org/10.7554/eLife.22689.044

Download elife-22689-fig5-data1-v2.xlsx

Figure 5—source data 2

Related to Figure 5—figure supplement 2E

https://doi.org/10.7554/eLife.22689.045

Download elife-22689-fig5-data2-v2.xlsx

Figure 5—source data 3

Related to Figure 5D.

https://doi.org/10.7554/eLife.22689.046

Download elife-22689-fig5-data3-v2.xlsx

Video 19

Download asset

As interference with rim movement led to an accumulation of cells outside the RNE, and these cells eventually adopted RNE-like morphology, we asked whether they would also adopt RNE fate. To test this, we used Vsx2, an early retinal transcription factor specific for RNE fate (Kimura et al., 2006). Time-lapse imaging of the transgenic line Tg(vsx2::GFP) expressing GFP under the vsx2 promoter allowed us to follow RNE specification. In control embryos, GFP was primarily expressed in the invaginating RNE layer and the signal was weak in cells within the rim zone. Only when rim cells reached the inner RNE did they express GFP more prominently (Video 20, Figure 6A). However, in the ezrin knockdown condition, rim cells that accumulated outside the RNE expressed GFP ectopically at this location (Video 21, Figure 6A,B). Such ectopic expression of vsx2 outside the optic cup was also observed in laminin and opo morphants. Here, cells failed to migrate and an S-shaped cup was formed (Figure 6B). Thus, vsx2 expression and RNE fate while seemingly temporally controlled were independent of correct cell position. In some ezrin morphants, these displaced RNE cells initiated a second invagination zone (Figure 6C). These phenotypes persisted until 36hpf, when neurogenesis in the RNE is known to start (Weber et al., 2014), resulting in abnormal architecture of the optic cup (Figure 6D). Thus, our results suggest that rim migration functions as a mechanism to ensure that initially ‘misplaced’ prospective RNE cells are moved to their correct location before they adopt RNE fate, making it a crucial step for all further retinal development.

Video 20

Download asset

Figure 6

Download asset Open asset

Video 21

Download asset

The teleost optic vesicle is an epithelial bilayer that needs to rearrange to form the hemispheric optic cup hosting the RNE that give rise to all retinal neurons later in development. To make this rearrangement efficient over a short developmental time span, the tissue has adopted an interplay of cell shape changes supported by basal constriction of RNE cells (Nicolás-Pérez et al., 2016) and migration of rim cells. However, it seems that rim involution plays a dominant role in this process as it translocates a substantial number of cells into the RNE that later contribute to invagination through basal actomyosin contractility and tissue compaction. As a result, the phenotypes observed upon perturbed rim involution are more severe than when basal constriction is impaired. Similar combinations of epithelial rearrangements are at play during other morphogenesis phenomena that occur in a rather short developmental time frame. For instance, during zebrafish posterior lateral line development, timely coordination between the epithelial rearrangements of microlumen formation and primordium migration determines the spacing of rosettes and in turn the lateral line architecture (Durdu et al., 2014). Thus, to fully understand the dynamics of morphogenetic processes it is important to decipher how diverse cell behaviors and their relationship orchestrate tissue formation in different contexts. This is important for determining the common and differential mechanisms underlying organ formation.

Interestingly, all of the dynamic morphogenetic changes elucidated here occur at the basal side of the epithelium. In the rim region, basal cell dynamics are crucial for cell migration, whereas in the invaginating region they assist the formation of stable basal adhesion and basal constriction. Basal domain dynamics are also involved in the formation of the zebrafish midbrain-hindbrain boundary, where the ECM is important for basal shrinkage of the boundary cells (Gutzman et al., 2008), and for the generation of the Drosophila follicular epithelium, where cell-ECM adhesion plays a crucial role in follicle rotation and egg elongation (Haigo and Bilder, 2011). Furthermore, also studies on in vitro optic cup organoids show that laminin is a crucial ECM component for successful cup formation (Eiraku et al., 2011). However, most studies of established systems of epithelial morphogenesis such as gastrulation movements in Drosophila or vertebrate neurulation (Guillot and Lecuit, 2013) have focused on the dynamics of the apical domain. It will therefore be important to further explore the role of dynamics of the basal domain and its interaction with the ECM during morphogenesis in different developmental contexts.

During development, cells rarely exhibit archetypal epithelial or mesenchymal characteristics but usually a mixture of both. This has led to the emerging concept of a continuum between the mesenchymal and epithelial state, a topic that recently gathered increasing attention (Campbell and Casanova, 2016; Bernadskaya and Christiaen, 2016). Depending on the context and the morphogenetic processes, cells utilize different states along this continuum. Rim cells represent such an example where initially cells exhibit both epithelial (apicobasal polarity and cell-cell adhesion) and mesenchymal characteristics (directed protrusive activity and dynamic cell-matrix contacts). However, once the cells involute and integrate into the invaginating RNE they stop their protrusive activity, adhere stably to the ECM and attain the full epithelial state. Thus, rim cells traverse along the continuum and modulate their cell behavior in space and time to shape the organ precursor, the RNE. Furthermore, the spatiotemporal regulation of this transition seems important as a premature transition to stably attached morphology hinders further steps of organogenesis. Similar cellular transitions are also observed at earlier stages of eye development, during optic vesicle evagination. During this event, the eye field cells that seem of more mesenchymal nature change their morphology and acquire apicobasal polarity to form the optic vesicle (Ivanovitch et al., 2013; Bazin-Lopez et al., 2015). Thus, in teleosts, the process of RNE formation from eye field to hemispherical retina utilizes gradual transitions from the mesenchymal to epithelial state at different developmental stages to efficiently shape the RNE into a hemisphere. Further research is required to understand how cell behavior is modulated spatiotemporally. Such investigations will provide insights into the transition along the continuum of states from mesenchyme to epithelium.

We show that rim cells undergo active cell migration using dynamic lamellipodia and cell-ECM attachments while staying integrated within the epithelial sheet. Thus, rim cell migration is a previously uncharacterized form of collective epithelial cell migration. Lately various morphogenetic phenomena have been discovered to include collective epithelial migration, including the well-studied systems of border cell migration in Drosophila, branching morphogenesis of trachea in Drosophila and mammary gland development, wound healing in mammals or lateral epidermal movement during dorsal closure (Scarpa and Mayor, 2016; Friedl and Gilmour, 2009). Studies of these different systems have led to the emerging view that in each context, the cells exhibit different extent of epithelial and migratory characters, thus broadening the definition of collective epithelial migration (Campbell and Casanova, 2016). Interestingly, in contrast to other collective migration phenomena, such as border-cell migration in Drosophila, migration of posterior lateral line primordium in zebrafish or wound healing (Scarpa and Mayor, 2016; Friedl and Gilmour, 2009), rim cell migration occurs in a part of a continuous epithelial sheet and lacks specific leader cells. Nevertheless, even without leader cells, rim migration occurs in a directed manner and lamellipodia formation is highly biased towards the direction of movement. This molecular and cellular directionality could possibly emerge from the distribution of forces present in the developing optic cup and/or the spatial distribution of surrounding ECM. Future studies will need to investigate these factors to understand how they influence directionality of rim migration.

Our data show that collective rim migration is indispensable for shaping the RNE, the organ precursor that gives rise to the retina. Rim migration translocates prospective neuroepithelial cells that initially reside in the epithelial sheet outside the presumptive RNE to their correct location in the hemispheric RNE. However, when these cells do not reach the RNE in time, they change their morphology and adhere to the underlying ECM at ectopic positions generating a negative feedback, as these cells can never be translocated further. Thus, timely controlled rim cell migration is critical to coordinate spatial positioning of cells with the timing of neuroepithelial fate determination. Failure of rim movement leads to enduring defects in optic cup architecture. This is highlighted by the observation that, in severe cases, a secondary invagination zone can be formed. Our study thereby reveals that RNE fate is independent of the position of cells, suggesting that while signaling pathways and patterning molecules are very important for RNE fate specification (Fuhrmann, 2010), they cannot ensure proper positioning of cells if epithelial rearrangements and morphogenic movements are impaired.

The crosstalk between morphogenesis and developmental patterning is crucial for successful organ development. Such developmental pattern is not only reflected in gene expression differences but also affects the distinct cell behaviors during early eye development (Picker et al., 2009; Heermann et al., 2015). The dorsal or distal layer of the optic vesicle starts invagination whereas the ventral region contributes to rim involution, predominantly in the ventral-temporal part of the retina (Schmitt and Dowling, 1994; Heermann et al., 2015; Picker et al., 2009; Kwan et al., 2012). Consistent with this, the phenotypes we observed in rim perturbation conditions result in accumulation of rim cells specifically at the ventral-temporal side of the optic cup. In contrast, cells are much less affected at the dorsal and nasal side. Interestingly, the architecture of the teleost optic vesicle is different compared to that of higher vertebrates. Nevertheless, ventral-temporal cell movements have not only been observed in teleosts but also during Xenopus and chick RNE morphogenesis (Holt, 1980; Kwan et al., 2012). Furthermore, studies in mouse have indicated that the invagination of the dorsal and ventral RNE can be uncoupled (Molotkov et al., 2006). Therefore, we speculate that the specific cell behavior at play might depend on the species and the shape of the optic vesicle. However, the use of differential morphogenetic strategies for dorsal and ventral retinal development is conserved in all in vivo systems studied so far. It is possible that such ‘pattern’ of morphogenic strategies allows the segregation of cells that experience different environments. Consequently, this could help the cells to attain the RNE fate in a stepwise manner to prepare for the later retinal developmental programs. As it has been recently suggested that such patterning could be conserved during in vitro optic cup formation in organoid cultures (Hasegawa et al., 2016) it will in the future be very exciting to assess how far developmental and cell-behavioral patterning in retinal organoids recapitulates the in vivo scenario at different developmental stages.

Wild type strains (WT-AB RRID:ZIRC_ZL1, WT-TL RRID:ZIRC_ZL86) and transgenic lines Tg(actb1:GFP-utrCH), (Behrndt et al., 2012), Tg(actb2:mCherry-Hsa.UTRN) (Compagnon et al., 2014), Tg(actb1:myl12.1-EGFP) (Maître et al., 2012), Tg(actb1:HRAS-EGFP) vu119 (Cooper et al., 2005), Tg(vsx2::GFP) (Kimura et al., 2006), Tg(βactin:mKate2-ras) were used. Zebrafish were maintained and bred at 26.5°C. Embryos were raised at 28°C and then transferred to 21°C at around 80% epiboly to slow down development. At 8 ss embryos were transferred back and maintained henceforth at 28°C. All animal work was performed in accordance with European Union (EU) directive 2011/63/EU as well as the German Animal Welfare Act.

For morpholino-mediated knockdown of gene function, the following amounts of morpholinos were injected in the yolk at one-cell stage:

0.4 ng laminin α−1 MO 5’TCATCCTCATCTCCATCATCGCTCA3’(Pollard et al., 2006),

3.8 ng ojoplano MO 5’ggactcacccaTCAGAAATTCAGCC3’ (Martinez-Morales et al., 2009), 1.6 ng ezrin MO 5’GATGTAGATGCCGATTCCTCTCGTC3’ (Link et al., 2006)

2 ng p53 MO 5’GCGCCATTGCTTTGCAAGAATTG3’(Robu et al., 2007).

See Supplementary file 1 for the phenotypes spread and number of embryos.

For whole tissue labeling, RNA was injected at the one-cell stage. For mosaic labeling of cells, either DNA was injected at one-cell stage or RNA was injected in a single blastomere at 16- to 32-cell stage. RNA was synthesized using the Ambion mMessage mMachine kit and injected at 50–60 pg per embryo, whereas DNA was injected at 15 pg per embryo.

pCS2+mKate2-ras (Weber et al., 2014), pCS2+GFP-ras (kind gift from A. Oates), pCS2+Par3-GFP (Tawk et al., 2007), pCS2+GFP-UtrophinCH (Burkel et al., 2007), pCS2+DD-myl12b-GFP (Norden et al., 2009), pCS2+ EMTB-tdTomato (kind gift from D. Gilmour), βactin::mKate2-ras (Icha et al., 2016), pCS2+Paxillin-mKate, pCS2+ Integrinβ1b-mKate. (for cloning strategies see below).

Dechorionated embryos were incubated in the drug solutions made in E3 medium. Rockout and Aphidicolin were dissolved in DMSO. Hydroxyurea was dissolved in water. Rockout treatment was started around 13–14 ss and was used at 100 µM with 1% DMSO (by volume) in E3. During time-lapse imaging, Rockout was used at 125 µM. For inhibition of cell proliferation, embryos were treated with a mixture of 30 mM HU and 210 µM Aphidicolin. HU+Aphi treatment was started at 10ss. Equivalent amounts of DMSO were added as solvent control.

The donor embryos (GFP-ras positive) were co-injected with the p53 morpholino along with Opo or ezrin morpholino. Embryos at high to sphere stage were dechorionated and some tens of cells from the donor embryos were transferred into the animal pole of the acceptors (mCherry-UtrCH positive). Transplanted embryos were then transferred to E3 medium supplemented with 100 U penicillin and streptomycin.

Embryos were fixed with 4% PFA in PBS overnight at 4°C, followed by permeabilisation using PBT (PBS with 0.8% Triton X-100). To improve permeability, embryos were trypsinized on ice for 15 min, followed by blocking with 10% normal goat serum and incubated in the primary antibody mix with 1% NGS in PBT for 60 h at 4°C. After washing, embryos were incubated with secondary antibody mix with 1% NGS in PBT for 60 h. The embryos were mounted either in agarose or in 80% glycerol.

The following dilutions were used. Primary antibodies: 1:50 anti-phospho-myosin (RRID:AB_330248, Cell signaling 3671), 1:100 anti-laminin (RRID:AB_477163, Sigma L-9393), 1:100 anti-chondroitin sulphate CS-56 (RRID:AB_298176, Abcam ab11570), 1:100 anti-fibronectin (RRID:AB_476976, Sigma F3648), 1:500 anti-pH3 (RRID:AB_2295065, Abcam ab10543), 1:500 anti-tRFP (RRID:AB_2571743, Evrogen AB233), 1:100 anti-GFP (RRID:AB_94936, Milipore MAB3580), 1:200 anti-PKC ζ C-20 (RRID:AB_2300359, Santa Cruz sc-216).

Secondary antibodies and fluorescent markers: 1:500 Alexa Fluor 488 anti-Rabbit (RRID:AB_141708, Invitrogen A21206), 1:500 Alexa Fluor 568 anti-Rabbit (RRID:AB_2534017, Thermo Fischer Scientific A10042), 1:500 Alexa Fluor 647 anti-Rabbit (RRID:AB_141775, Invitrogen A21245), 1:500 Alexa Fluor 488 anti-mouse (RRID:AB_141606, Invitrogen A21200), Alexa Fluor 594 anti-mouse (RRID:AB_141630, Invitrogen A21201), 1:500 Alexa Fluor 647 anti-rat (RRID:AB_141778, Invitrogen A21247), 1:50 Alexa Fluor 488 Phalloidin (RRID:AB_2315147, Molecular Probes A12379), 1:50 Rhodamine-Phalloidin (RRID:AB_2572408, Molecular probes R415), DAPI.

Image analysis was performed using Fiji platform RRID:SCR_002285 (Schindelin et al., 2012).

The following constructs were generated using Gateway cloning system.

1 nl of the mix of Tol2 plasmid Tol2-bactin::mKate2-ras (Icha et al., 2016) containing 20 ng/μl and Tol2 transposase RNA 30 ng/μl in ddH₂O was injected into the cytoplasm of one-cell stage embryos. F₀ embryos with observed fluorescence signal were grown to adulthood and Tg carriers were identified by outcross with wild type fish.

For each condition protein sample from five embryos was used and western blot performed. Following antibodies dilutions were used: 1:500 anti-laminin (RRID:AB_477163, Sigma L-9393), 1:500 anti-phospho-ERM (RRID:AB_2262427, Cell Signaling 3141), 1:10000 Anti-α-Tubulin (RRID:AB_477582, Sigma T6074), 1:20000 peroxidase conjugated anti-Mouse (RRID:AB_2340061, Jackson Immuno 315-035-003), 1:20000 peroxidase conjugated anti-Rabbit (RRID:AB_2313567, Jackson Immuno 111-035-003).

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

We thank Y Bellaiche, E Knust, J Martinez-Morales, D Mateju, M Palfy, and IK Patten for helpful comments on the manuscript. We also thank the Norden lab for useful discussions, B Lombardot and R Haase for the script allowing analysis of actomyosin distribution. Specially N Iffländer as well as C Fröb, J Icha, S Kaufmann, M Palfy, I Yanakieva, the light microscopy and fish facility are thanked for experimental help. Thanks to D Gilmour, C Henry and A Oates for sharing constructs.

JS is a member of the IMPRS-CellDevoSys PhD program and supported by a fellowship by the DIGS-BB. CN was supported by the Human Frontier Science Program (CDA-00007/2011) and the German Research Foundation (DFG). The authors declare no conflict of interest.

Animal experimentation: All animal work was performed in accordance with European Union (EU) directive 2011/63/EU as well as the German Animal Welfare Act.

© 2017, Sidhaye et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.