The skin of Xenopus embryos contains numerous multiciliated cells (MCCs), which collectively generate a directed fluid flow across the epithelial surface essential for distributing the overlaying mucous. MCCs develop into highly specialized cells to generate this flow, containing approximately 150 evenly spaced centrioles that give rise to motile cilia. MCC-driven fluid flow can be impaired when ciliary dysfunction occurs, resulting in primary ciliary dyskinesia (PCD) in humans. Mutations in a large number of genes (∼50) have been found to be causative to PCD. Recently, studies have linked low levels of Adenylate Kinase 7 (AK7) gene expression to patients with PCD; however, the mechanism for this link remains unclear. Additionally, AK7 mutations have been linked to multiple PCD patients. Adenylate kinases modulate ATP production and consumption, with AK7 explicitly associated with motile cilia. Here we reproduce an AK7 PCD-like phenotype in Xenopus and describe the cellular consequences that occur with manipulation of AK7 levels. We show that AK7 localizes throughout the cilia in a DPY30 domain-dependent manner, suggesting a ciliary function. Additionally, we find that AK7 overexpression increases centriole number, suggesting a role in regulating centriole biogenesis. We find that in AK7-depleted embryos, cilia number, length, and beat frequency are all reduced, which in turn significantly decreases the tissue-wide mucociliary flow. Additionally, we find a decrease in centriole number and an increase in sub-apical centrioles, implying that AK7 influences both centriole biogenesis and docking, which we propose underlie its defect in ciliogenesis. We find that both the AK domain and the DPY30 domain are required for proper centriole regulation. We propose that AK7 plays a role in PCD by impacting centriole biogenesis and apical docking, ultimately leading to ciliogenesis defects that impair mucociliary clearance.

Tissue crowding represents a critical challenge to epithelial tissues, which often respond via the irreversible process of live cell extrusion. We report that apical size reduction via macropinocytosis serves as a malleable and less destructive form of tissue remodeling that can alleviate the need for cell loss. We find that macropinocytosis is triggered by tissue crowding via mechanosensory signaling, leading to substantial internalization of apical membrane. This drives a reduction in apical surface which alleviates crowding. We report that this mechanism regulates the long-term organization of the developing epithelium and controls the timing of proliferation-induced cell extrusion. Additionally, we observe a wave of macropinocytosis in response to acute external compression. In both scenarios, inhibiting macropinocytosis induces a dramatic increase in cell extrusion suggesting cooperation between cell extrusion and macropinocytosis in response to both developmental and external compression. Our findings implicate macropinocytosis as an important regulator of dynamic epithelial remodeling.

Xenopus embryos are covered with a complex epithelium containing numerous multiciliated cells (MCCs). During late-stage development, there is a dramatic remodeling of the epithelium that involves the complete loss of MCCs. Cell extrusion is a well-characterized process for driving cell loss while maintaining epithelial barrier function. Normal cell extrusion is typically unidirectional, whereas bidirectional extrusion is often associated with disease (e.g. cancer). We describe two distinct mechanisms for MCC extrusion, a basal extrusion driven by Notch signaling and an apical extrusion driven by Piezo1. Early in the process there is a strong bias towards basal extrusion, but as development continues there is a shift towards apical extrusion. Importantly, response to the Notch signal is age dependent and governed by the maintenance of the MCC transcriptional program such that extension of this program is protective against cell loss. In contrast, later apical extrusion is regulated by Piezo1, such that premature activation of Piezo1 leads to early extrusion while blocking Piezo1 leads to MCC maintenance. Distinct mechanisms for MCC loss underlie the importance of their removal during epithelial remodeling.

Stereocilia are actin-based projections of hair cells that are arranged in a step like array, in rows of increasing height, and that constitute the mechanosensory organelle used for the senses of hearing and balance. In order to function properly, stereocilia must attain precise sizes in different hair cell types and must coordinately form distinct rows with varying lengths. Espins are actin-bundling proteins that have a well-characterized role in stereocilia formation; loss of function mutations in Espin result in shorter stereocilia and deafness in the jerker mouse. Here we describe the generation of an Espin overexpressing transgenic mouse line that results in longer first row stereocilia and discoordination of second-row stereocilia length. Furthermore, Espin overexpression results in the misregulation of other stereocilia factors including GNAI3, GPSM2, EPS8, WHRN, and MYO15A, revealing that GNAI3 and GPSM2 are dispensable for stereocilia overgrowth. Finally, using an in vitro actin polymerization assay we show that espin provides an anti-capping function that requires both the G-actin binding WH2 domain as well as either the C-terminal F-actin binding domain or the internal xAB actin-binding domain. Our results provide a novel function for Espins at the barbed ends of actin filaments distinct from its previous known function of actin bundling that may account for their effects on stereocilia growth.

Centriole amplification in multiciliated cells occurs in a pseudo-cell cycle regulated process that typically utilizes a poorly characterized molecularly dense structure called the deuterosome. We identified the centrosomal protein Cep70 as a novel deuterosome-associated protein that forms a complex with other deuterosome proteins, CCDC78 and Deup1. Cep70 dynamically associates with deuterosomes during centriole amplification in the ciliated epithelia of Xenopus embryos. Cep70 is not found in nascent deuterosomes prior to amplification. However, it becomes localized at deuterosomes at the onset of centriole biogenesis and remains there after the completion of centriole amplification. Deuterosome localization requires a conserved C-terminal "Cep70" motif. Depletion of Cep70 using morpholino oligos or CRISPR/Cas9 editing in F0 embryos leads to a severe decrease in centriole formation in both endogenous MCCs, as well as ectopically induced MCCs. Consistent with a decrease in centrioles, endogenous MCCs have defects in the process of radial intercalation. We propose that Cep70 represents a novel regulator of centriole biogenesis in MCCs.

Scribble, a member of the LAP protein family, contributes to the apicobasal polarity (ABP) of epithelial cells. The LAP-unique region of these proteins, which is essential and sufficient for ABP, includes a conserved Leucine-Rich Repeat (LRR) domain. The major binding partners of this region that could regulate ABP remain unknown. Here, using proteomics, native gel electrophoresis, and site-directed mutagenesis, we show that the concave surface of LRR domain in Scribble participates in three types of mutually exclusive interactions-(i) homodimerization, serving as an auto-inhibitory mechanism; (ii) interactions with a diverse set of polarity proteins, such as Llgl1, Llgl2, EPB41L2, and EPB41L5, which produce distinct multiprotein complexes; and (iii) a direct interaction with the protein phosphatase, PP1. Analogy with the complex between PP1 and LRR domain of SDS22, a well-studied PP1 regulator, suggests that the Scibble-PP1 complex stores a latent form of PP1 in the basolateral cell cortex. Such organization may generate a dynamic signaling network wherein PP1 could be dispatched from the complex with Scribble to particular protein ligands, achieving fast dephosphorylation kinetics.

Multiciliated cells (MCCs) are extremely highly differentiated, presenting >100 cilia and basal bodies. Therefore, MCC fate is thought to be terminal and irreversible. We analyzed how MCCs are removed from the airway-like mucociliary Xenopus epidermis during developmental tissue remodeling. We found that a subset of MCCs undergoes lateral line-induced apoptosis, but that the majority coordinately trans-differentiate into goblet secretory cells. Both processes are dependent on Notch signaling, while the cellular response to Notch is modulated by Jak/STAT, thyroid hormone, and mTOR signaling. At the cellular level, trans-differentiation is executed through the loss of ciliary gene expression, including foxj1 and pcm1, altered proteostasis, cilia retraction, basal body elimination, as well as the initiation of mucus production and secretion. Our work describes two modes for MCC loss during vertebrate development, the signaling regulation of these processes, and demonstrates that even cells with extreme differentiation features can undergo direct fate conversion.

The epidermis of the Xenopus embryo has emerged as a powerful tool for studying the development of a ciliated epithelium. Interspersed throughout the epithelium are multiciliated cells (MCCs) with 100+ motile cilia that beat in a coordinated manner to generate fluid flow over the surface of the cell. MCCs are essential for various developmental processes and, furthermore, ciliary dysfunction is associated with numerous pathologies. Therefore, understanding the cellular mechanisms involved in establishing a ciliated epithelium are of particular interest. MCCs originate in the inner epithelial layer of Xenopus skin, where Notch signaling plays a critical role in determining which progenitors will adopt a ciliated cell fate. Then, activation of various transcriptional regulators, such as GemC1 and MCIDAS, initiate the MCC transcriptional program, resulting in centriole amplification and the formation of motile cilia. Following specification and differentiation, MCCs undergo the process of radial intercalation, where cells apically migrate from the inner layer to the outer epithelial layer. This process involves the cooperation of various cytoskeletal networks, activation of various signaling molecules, and changes in cell-ECM and cell-cell adhesion. Coordination of these cellular processes is required for complete incorporation into the outer epithelial layer and generation of a functional ciliated epithelium. Here, we highlight recent advances made in understanding the transcriptional cascades required for MCC specification and differentiation and the coordination of cellular processes that facilitate radial intercalation. Proper regulation of these signaling pathways and processes are the foundation for developing a ciliated epithelium.

Purpose: To understand the relationship between ciliogenesis and autophagy in the corneal epithelium. Methods: siRNAs for EphA2 or PLD1 were used to inhibit protein expression in vitro. Morpholino-anti-EphA2 was used to knockdown EphA2 in Xenopus skin. An EphA2 knockout mouse was used to conduct loss of function studies. Autophagic vacuoles were visualized by contrast light microscopy. Autophagy flux, was measured by LC3 turnover and p62 protein levels. Immunostaining and confocal microscopy were conducted to visualize cilia in cultured cells and in vivo. Results: Loss of EphA2 (i) increased corneal epithelial thickness by elevating proliferative potential in wing cells, (ii) reduced the number of ciliated cells, (iii) increased large hollow vacuoles, that could be rescued by BafA1; (iv) inhibited autophagy flux and (v) increased GFP-LC3 puncta in the mouse corneal epithelium. This indicated a role for EphA2 in stratified epithelial assembly via regulation of proliferation as well as a positive role in both ciliogenesis and end-stage autophagy. Inhibition of PLD1, an EphA2 interacting protein that is a critical regulator of end-stage autophagy, reversed the accumulation of vacuoles, and the reduction in the number of ciliated cells due to EphA2 depletion, suggesting EphA2 regulation of both end-stage autophagy and ciliogenesis via PLD1. PLD1 mediated rescue of ciliogenesis by EphA2 depletion was blocked by BafA1, placing autophagy between EphA2 signaling and regulation of ciliogenesis. Conclusion: Our findings demonstrate a novel role for EphA2 in regulating both autophagy and ciliogenesis, processes that are essential for proper corneal epithelial homeostasis.

Article Figures and data Abstract Introduction Results and discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract How cells count and regulate organelle number is a fundamental question in cell biology. For example, most cells restrict centrioles to two in number and assemble one cilium; however, multiciliated cells (MCCs) synthesize hundreds of centrioles to assemble multiple cilia. Aberration in centriole/cilia number impairs MCC function and can lead to pathological outcomes. Yet how MCCs control centriole number remains unknown. Using Xenopus, we demonstrate that centriole number scales with apical area over a remarkable 40-fold change in size. We find that tensile forces that shape the apical area also trigger centriole amplification based on both cell stretching experiments and disruption of embryonic elongation. Unexpectedly, Piezo1, a mechanosensitive ion channel, localizes near each centriole suggesting a potential role in centriole amplification. Indeed, depletion of Piezo1 affects centriole amplification and disrupts its correlation with the apical area in a tension-dependent manner. Thus, mechanical forces calibrate cilia/centriole number to the MCC apical area via Piezo1. Our results provide new perspectives to study organelle number control essential for optimal cell function. Introduction Organelles compartmentalize cells into discrete functioning units. Cells must regulate the number of organelles to achieve proper function (Marshall, 2007; Marshall, 2016; Nigg and Holland, 2018; Rafelski and Marshall, 2008). For example, multiciliated cells (MCCs) line the epithelia of the brain ventricles, the airway, and the oviduct where motile cilia propel extracellular fluid to circulate cerebrospinal fluid, remove pathogens, and move the ova (Spassky and Meunier, 2017). Depending on the location, MCCs synthesize between 30 and 300 motile cilia (Spassky and Meunier, 2017). Assembly of too few or too many cilia impairs MCC function and is associated with several diseases including Primary Ciliary Dyskinesia, suggesting the existence of an active mechanism that controls cilia number (Boon et al., 2014; Spassky and Meunier, 2017; Wallmeier et al., 2014). Yet, the cellular and molecular mechanisms that control the number of cilia in MCCs remain unknown. To shed light on mechanisms, we used the Xenopus embryonic epidermis, an established, versatile, in vivo model to study MCCs (Walentek and Quigley, 2017; Werner and Mitchell, 2013). There, MCCs are first specified in the basal epithelia, where they begin to synthesize centrioles using specialized structures called deuterosomes (Figure 1a, Step 0) (Klos Dehring et al., 2013; Zhao et al., 2013). As MCCs intercalate into the outer epithelial cell layer and expand their apical surface, centrioles migrate apically, dock at the apical surface, and provide the platform for assembly of motile cilia (Figure 1a, Steps 1–4) (Deblandre et al., 1999; Kulkarni et al., 2018a; Stubbs et al., 2006; Zhang and Mitchell, 2015). As such, in this study, we focused our efforts on counting centrioles as a simple, efficient proxy for cilia number using chibby-GFP, a marker for mature centrioles (Burke et al., 2014). Figure 1 with 2 supplements see all Download asset Open asset Centriole number scales with apical area in Xenopus MCCs. (a) Schematic representing the current understanding of how MCCs differentiate and develop (Steps 0–4). Xenopus embryonic development is closely linked (dashed arrows) to MCC development. (b) MCCs captured in different stages of development (Steps 2–4) and labeled with chibby-GFP (centrioles). Dotted line represents the cell boundary. (c) Regression plot showing the positive correlation between apical area and number of centrioles at the apical surface in developing (green, Step 2, 3) and mature MCCs (blue, Step 4). (d) One-cell stage and stage 28 embryos of Xenopus tropicalis and Xenopus laevis. Images are to scale. Mature (Step 4) epidermal MCCs marked with chibby-GFP (centrioles) and phalloidin (F-actin) of (e) X. tropicalis and (f) X. laevis embryos at stage 28. Quantitation of (g) apical area and (h) centriole number in MCCs of X. tropicalis and X. laevis. The statistical comparison between the treatments is done using an unpaired t test. (i) Regression plot showing the positive correlation between apical area and centriole number across species in mature MCCs. (j) Regression plot showing the scaling relationship exists over a 40-fold change in apical area among different treatments. R2 is the correlation coefficient. * indicates statistical significance at p < 0.05. n = number of cells from 15 to 25 embryos. The data is uploaded as source data 1. Figure 1—source data 1 Source data related to Figure 1 . https://cdn.elifesciences.org/articles/66076/elife-66076-fig1-data1-v2.xlsx Download elife-66076-fig1-data1-v2.xlsx Results and discussion MCC development is closely linked to embryonic development (Figure 1a). Therefore, we collected embryos at different developmental stages (from stage 20 to stage 28) to examine MCCs at various stages of apical expansion, ranging from MCCs that have just intercalated (Step 2) to fully mature MCCs (step 4). We measured the number of centrioles at the apical surface and the apical area. Surprisingly, we observed a strong correlation between the apical area and the number of centrioles at the apical surface (Figure 1b,c). Next, we wanted to test if this relationship would persist if the MCC apical area became larger. We employed four different approaches to increase the apical area of MCCs. First, two species of Xenopus are common models for cell biology (X. laevis and X. tropicalis), and due to the evolutionary variation in embryonic sizes of the two species, they are useful for scaling experiments (Figure 1d – compare relative sizes of the eggs and embryo) (Levy and Heald, 2012). Compared to X. tropicalis, the X. laevis embryo is larger with significantly larger MCCs (Figure 1d–g, median ± SD, 391 ± 86 μm2 vs. 270 ± 53 μm2 in X. tropicalis). X. laevis MCCs also have significantly more centrioles (Figure 1h, median ± SD, 195 ± 27 compared to 150 ± 20 in X. tropicalis). Interestingly, by combining data from both species, we observed a clear trend where centriole number scales with MCC apical area (Figure 1i). Second, in X. laevis, we converted epithelial goblet cells (which normally secrete mucus) to MCCs by overexpressing the master regulator of multiciliogenesis, mcidas (multiciliate differentiation and DNA synthesis-associated cell cycle protein) (Figure 1—figure supplement 1a; Stubbs et al., 2012). These induced MCCs in X. laevis are larger than X. tropicalis MCCs and have proportionately more centrioles (Figure 1—figure supplement 1b,c). Interestingly, the apical area of mature MCCs and the induced MCCs of X. laevis were similar and so were the number of centrioles (Figure 1—figure supplement 1c). Third, we induced cytokinesis defects by knocking down ccdc11 using a morpholino oligo (MO) in X. tropicalis (Kulkarni et al., 2018b). MCCs are mitotically mature; however, if their progenitor fails to undergo cytokinesis, then the resultant MCC can be much larger (Figure 1—figure supplement 2a). With this strategy, we identified MCCs with significantly larger apical areas (median ± SD, 437 ± 154 μm2 vs. 202 ± 38 μm2 in controls) (Figure 1—figure supplement 2a,b), and these cells had proportionately more centrioles (median ± SD, 227 ± 74 vs. 138 ± 26 in controls) (Figure 1—figure supplement 2c,d). Finally, in X. laevis, we fused MCCs with neighboring (most likely) non-MCCs (confirmed by the presence of two nuclei – dashed lines in Figure 1—figure supplement 2e), resulting in much larger cells with increased apical area (median ± SD, apical area of 534 ± 149 μm2 vs. 337 ± 49 μm2 in controls) and more centrioles (median ± SD, 235 ± 40 vs. 168 ± 21 in controls) (Figure 1—figure supplement 2e–h). Interestingly, in each experiment, centriole number increased in proportion to the apical area suggesting that centriole amplification is a plastic process and cells can calibrate centriole number in response to cell size perturbations. By combining the data from controls and manipulated embryos, we found that this scaling relationship could be observed over a 40-fold change in the apical area, with the smallest apical area being ~25 μm2 and the largest about 1000 μm2 (Figure 1j). However, these experiments are limited in two ways. First, in our experiments, we increased the entire volume of the cell not just the apical area. Additionally, in cells with a cytokinesis defect or cell-cell fusion, we have combined two or more cells leading to an increase in the centriole number. Nevertheless, despite these limitations, the correlation between apical area and centriole number appears robust, and in subsequent experiments, we strived to overcome these limitations. While these experiments suggest that the apical area of an MCC may regulate centriole number, we sought to test the alternative hypothesis that centriole number may determine apical area or that there may be a feedback mechanism between centriole number and apical area. We can manipulate centriole number in two ways: increase the number of centrioles by overexpressing cep152 (Collins et al., 2020; Klos Dehring et al., 2013) or decrease the number of centrioles with Centrinone treatment, a PLK4 inhibitor (Wong et al., 2015). We first increased the centriole number by overexpressing cep152 in X. laevis (Figure 2a), which increased the number of centrioles (median ± SD, 467 ± 113 vs. 160 ± 27 in controls) and was accompanied by a correlated increase in the apical area (median ± SD, 692 ± 296 μm2 vs. 236 ± 58 μm2 in controls) (Figure 2a–d). Next, we reduced centriole numbers with Centrinone. Centriole synthesis begins during intercalation, when the cells are in the basal layer (Figure 1a, Step 0). To allow the chemical inhibitor to access the cells in the basal layer, we generated Xenopus embryonic ‘stem cell’ explants (commonly referred to as animal caps), which auto-differentiate into an embryonic epidermis replete with MCCs (Figure 2e). We harvested animal caps from X. tropicalis embryos and grew them on fibronectin-coated slides with exposure to Centrinone or vehicle alone until control embryos reached stage 25–26 (Figure 2e,f). We successfully reduced the median number of centrioles from 104 in controls to 25 in Centrinone-treated MCCs (Figure 2f,g). Despite a dramatic reduction in the number of centrioles, we found a slight increase in the apical area of Centrinone-treated MCCs as compared to controls (median ± SD, 175 ± 67 vs. 141 ± 38 μm2 in controls) (Figure 2h). From this result, we conclude that a minimum apical size can be achieved independent of the centriole amplification (Figure 2f–i). Once this minimum size is reached, then centrioles may contribute to apical expansion (Figure 2a–d). Moving forward, we focused on the hypothesis that the apical area may fine tune centriole number. Figure 2 Download asset Open asset Perturbation of centriole amplification affects apical area contextually. (a) Mature (Step 4) epidermal MCCs marked with chibby-GFP (centrioles, green), and phalloidin (F-actin, magenta) in control and Cep152 overexpressed (OE) embryos. Quantitation of (b) apical area and (c) centriole number in MCCs of control and CEP 152 OE embryos at stage 28. (d) Regression plot showing the positive correlation between apical area and centriole number. (e) Experimental design to block centriole amplification in MCCs using Centrinone in animal caps. We dissected the animal caps at stage 8–9 and tethered them to slides using fibronectin. At stage 14 (based on unmanipulated sibling embryos), we exposed the caps to Centrinone until their unmanipulated sibling embryos reached stage 25–26. (F) Epidermal MCCs marked with chibby-GFP (centrioles, green), and phalloidin (F-actin, magenta) in control and Centrinone-treated animal caps. Quantitation of (g) apical area and (h) centriole number in MCCs of control and Centrinone-treated animal caps. (i) Regression plot showing the loss of correlation between apical area and centriole number in Centrinone-treated MCCs. * indicates statistical significance at p < 0.05. The statistical comparison between the treatments (b, c, g, h) is done using an unpaired t test. R2 is the correlation coefficient. n = number of cells collected from 10 to 15 embryos. The data is uploaded as source data 2. Figure 2—source data 1 Source data related to Figure 2. https://cdn.elifesciences.org/articles/66076/elife-66076-fig2-data1-v2.xlsx Download elife-66076-fig2-data1-v2.xlsx While we initially focused on the number of centrioles at the apical surface, this may not reflect the total number of centrioles in the cell. Previous studies have noted the presence of centrioles in the intercalating MCCs (Figure 1a, Steps 0–1), but the number of these centrioles is unknown (Collins et al., 2020; Werner et al., 2014). One possibility is that MCCs assemble all of the centrioles (~150 in X. tropicalis) during intercalation with subsequent waves of either synthesis or degradation depending on the final apical area. Alternatively, MCCs may continuously produce centrioles and halt this process based on the final apical area. In either model, MCCs would require a cellular and molecular mechanism to measure the apical area. To differentiate between these hypotheses, we set out to image centrioles in the intercalating cells. To image deep within the cytoplasm, we used animal caps which are relatively transparent compared to whole embryos and imaged centrioles (Chibby-GFP) along the apico-basal axis of intercalating MCCs. Using segmentation and 3D reconstruction, we could observe the process of centriole migration and count all the centrioles in the cytoplasm of intercalating MCCs (presumptive MCC border marked by white dotted line, Figure 3a–d, Videos 1 and 2). In these intercalating cells, the number of centrioles was 75 (median), approximately half the number in mature X. tropicalis MCCs (Figure 3e, blue, Step 4; magenta, Steps 0–1). Interestingly, X. laevis MCCs also make 90 (median) centrioles during intercalation, again half the number of centrioles in mature MCCs (Figure 3f blue, Step 4; magenta, Steps 0–1). From these results, we conclude that (1) Xenopus MCCs synthesize half the total number of centrioles prior to intercalation, (2) these centrioles dock to the apical surface in a manner that scales with apical area, and (3) the remaining half of the number of centrioles must be synthesized in a second round that is regulated based on the apical area. Figure 3 Download asset Open asset MCCs synthesize about half the total number of centrioles during intercalation. (a–c) Centrioles (chibby-GFP, green) and F-actin (phalloidin, magenta) in intercalating MCCs (Step 1). Dotted while lines show the border of the intercalating MCCs. (d) The same MCC is segmented using IMARIS to show individual centrioles in grey and F-actin in magenta. MCCs generate half of the total number of centrioles just prior to intercalation in (e) X. tropicalis and (f) X. laevis. * indicates statistical significance at p < 0.05. The statistical comparison between the treatments is done using an unpaired t test. n = number of cells from 10 to 15 embryos/species. The data is uploaded as source data 3. Figure 3—source data 1 Source data related to Figure 3. https://cdn.elifesciences.org/articles/66076/elife-66076-fig3-data1-v2.xlsx Download elife-66076-fig3-data1-v2.xlsx Video 1 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg Centrioles (green) and F-actin (magenta). Centrioles dispersed below the apical surface of an intercalating MCC. Video 2 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg Segmentation and 3D reconstruction of the Video 1 using IMARIS. The apical area of the cell is dependent on multiple parameters including but not limited to the overall size of the cell, cell autonomous pushing forces, as well as pulling forces by neighboring cells (Guillot and Lecuit, 2013; Heisenberg and Bellaïche, 2013; Mao and Baum, 2015; Sedzinski et al., 2016). Specifically, during intercalation, the MCC is thought to cell autonomously push against its neighbors to expand its apical surface. Subsequently, neighboring cells pull on the MCC at cell junctions, expanding the apical area further (Sedzinski et al., 2016). To elucidate the contributions of pushing vs. pulling forces, we decided to examine the shape of the cells during apical expansion. Cell autonomous pushing forces would be radially symmetric so the apical surface should expand circularly (Sedzinski et al., 2016). On the other hand, cell non-autonomous pulling forces would depend on the relative positions of the neighboring cells and cell junctions leading to a polygonal apical shape (Sedzinski et al., 2016). The thinness ratio (TR) which relates the area of a shape to the square of its perimeter can detect these changes in cell shapes (Figure 4a). The TR is 1 for a circle and < 1 for polygons (Figure 4a). When we plotted the TR as a function of apical area, we found that MCCs with small apical areas have a TR of nearly 1, while the TR decreases to 0.8 as the apical area increases to ~300 μm2 (Figure 4b,c, Video 3). Therefore, TR measurements support the notion that the initial apical expansion is driven by cell autonomous pushing forces, while subsequent apical expansion is driven largely by cell non-autonomous pulling forces. Figure 4 with 3 supplements see all Download asset Open asset Mechanical stretch triggers centriole amplification in MCCs. (a) Schematic showing the effect of cell autonomous pushing (blue) vs. cell non-autonomous pulling forces (red) on cell shape and the thinness ratio (TR). (b) A single MCC (marked by membrane-RFP) undergoing apical expansion and in the process changing the cell shape from circular (TR=0.93) to more polygonal (TR=0.78). (c) A regression plot showing the negative correlation between the TR and the apical area. Magenta: 100–150 μm2, Green: 151–250 μm2, Blue: 251–350 μm2, Red: 351–600 μm2. n = 122 cells collected from 20 to 25 embryos (d) Binning the apical area shows that the increase in apical area leads to a significant reduction in the TR. MCCs marked with chibby-GFP (centrioles, green), and phalloidin (F-actin, magenta) in (e) control embryos, (f) untethered animal caps, (g) tethered animal caps, and (h) mechanically stretched animal caps. The statistical comparison between the treatments is done one-way ANOVA test followed by Tukey’s multiple comparisons test. Quantitation of (i) apical area and (j) centriole number in MCCs of animal caps subjected to different mechanical stimuli. Dashed line indicates the median value of controls. * indicates statistical significance at p < 0. 05. The statistical comparison between the treatments is done using the Brown-Forsythe and Welch ANOVA test followed by the Dunnett's T3 multiple comparisons test. n = number of cells. Data for untethered and tethered caps was collected from 10 to 12 animal caps. Data for stretched animal caps was collected from six to nine animal caps. (k) Regression plot demonstrates the scaling relation between the apical area and centriole number across different treatments. The data is uploaded as source data 4. Figure 4—source data 1 Source data related to Figure 4. https://cdn.elifesciences.org/articles/66076/elife-66076-fig4-data1-v2.xlsx Download elife-66076-fig4-data1-v2.xlsx Video 3 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg MCC labeled with membrane-RFP undergoing expansion of its apical surface. We speculated that cell non-autonomous pulling forces that drive the final phase of apical expansion might define the apical area and centriole number in MCCs. To test the hypothesis, we began with an embryological approach to manipulate the apical area. In a developing embryo, morphogenetic movements create forces that lead to dramatic shape changes that transform a spherical embryo (stage 9–10) to an elongated one (stage 28), presumably, exerting stretching forces on the epidermal MCCs to increase their apical area (Figure 1a). For example, Spemann’s Organizer, which is dependent on Wnt signaling, dorsalizes the mesoderm and ectoderm, which subsequently creates considerable elongation forces (De Robertis et al., 2000; Harland and Gerhart, 1997; Hikasa and Sokol, 2013; Keller and Sutherland, 2020; Kiecker, 2000). By depleting β-catenin, a key effector of the Wnt signaling pathway, we can eliminate the formation of Spemann’s Organizer and generate cylindrically symmetric embryos that lack dorsal structures and have much less elongation compared to control embryos (Figure 4—figure supplement 1a,c, Heasman et al., 1994; Khokha et al., 2005). While β-catenin-depleted embryos can form functioning MCCs that generate fluid flow, both the MCCs (median ± SD, 106 ± 26 μm2 vs. 267 ± 64 μm2 in controls) and non-MCCs (median ± SD, 319 ± 104 μm2 vs. 428 ± 125 μm2 in controls,) have smaller apical areas (Figure 4—figure supplement 1b,d–f). Interestingly, the centriole number in these embryos is also significantly decreased (median ± SD, 100 ± 16 vs. 148 ± 26 in controls, Figure 4—figure supplement 1g,h), approaching the 75 centrioles formed prior to intercalation. Further, the TR in β-catenin-depleted MCCs is significantly higher and closer to 1 (median ± SD, 0.93 ± 0.05 vs. 0.77 ± 0.04 in controls, Figure 4—figure supplement 1i,j), supportive of a significant reduction of pulling forces exerted on MCCs. This result suggests that the lack of embryonic elongation forces in β-catenin-depleted embryos causes the reduction in the MCC apical area and centriole number. However, a challenge in this experiment is the confounding effects generated by genetic manipulations, such as diminished Wnt signaling or potential changes in cell adhesion in β-catenin depleted embryos. To avoid these confounding effects, we sought to manipulate MCCs using non-genetic tools. We returned to animal caps, Xenopus stem cell explants that auto-differentiate into an embryonic multiciliated epidermis and raised them in two different conditions. In the first condition, we harvested animal caps and cultured them on agarose. In this case, because the cells do not adhere to agarose, the animal caps roll up to form irregular spherical structures which we called ‘untethered’ explants (Figure 4—figure supplement 2a,b, Video 4). The MCCs in these explants had an apical area just slightly larger than in β-catenin-depleted embryos (median ± SD, 116 ± 37 μm2 compared to 106 ± 26 μm2 in β-catenin-depleted embryos and 263 ± 46 μm2 in controls, Figure 4e,f,i) and a correlated decrease in centriole number (median ± SD, 105 ± 27 compared to 100 ± 16 in β-catenin-depleted embryos and 149 ± 14 in controls, Figure 4j). In the second condition, we harvested animal caps and cultured them on fibronectin-coated slides. In this case, the cells adhere to the slide and spread outward (Stepien et al., 2019). As a result, these ‘tethered’ explants are stretched along the slide to form flat epithelia (Figure 4—figure supplement 2c,d, Video 5; Stepien et al., 2019). In these tethered explants, the apical area of both non-MCCs (Figure 4—figure supplement 3, median ± SD, 372 ± 159 μm2 compared to 170 ± 86 μm2 in untethered caps and 465 ± 177 μm2 in controls) and the MCCs (Figure 4e–g,i, median ± SD, 210 ± 43 μm2 compared to 116 ± 37 μm2 in untethered caps compared to 263 ± 46 μm2 in controls) are increased compared to the untethered caps but are slightly smaller than epithelial cells in the embryo suggesting that additional forces or factors in the embryo may contribute to the apical area. Nevertheless, the tethered caps had a significant increase in the number of centrioles (Figure 4j, median ± SD, 130 ± 25 vs 105 ± 27 in untethered caps compared to 149 ± 14 in controls) in an area-dependent manner. Video 4 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg Untethered cap forms an irregular spherical structure. F-actin is in magenta. Video 5 Download asset This video cannot be played in place because your browser does support HTML5 video. You may still download the video for offline viewing. Download as MPEG-4 Download as WebM Download as Ogg Tethered animal cap forms a flat multiciliated epithelium. F-actin is in magenta. To understand the contribution of pulling forces in defining the apical area, we analyzed cell shapes and measured the TR. Specifically, by binning the data based on our results, from 0 to 100 μm2 (apical areas of MCCs in the initial stages of development, step 2, median ± SD, TR: 0.90 ± 0.05), 100–150 μm2 (untethered caps, median ± SD, TR: 0.89 ± 0.05), 150–250 μm2 (tethered caps, median ± SD, TR: 0.79 ± 0.06), 250–350 μm2 (wildtype X. tropicalis MCCs, median ± SD, TR: 0.79 ± 0.06), the TR reduces significantly as the cells become larger, supporting the increasing contribution of pulling forces on defining the apical area (Figure 4d). Taken together, these results suggested that tension generated by stretching within the epithelial sheet is critical to achieve proper apical area and triggers centriole amplification over the initial set of 75 centrioles. To directly test the role of stretching force on centriole number, we applied an artificial radial stretch to the explants. Specifically, we raised X. tropicalis explants on a silicone membrane coated with fibronectin until sibling control embryos reached stage 26. At this stage, MCCs are nearly mature, and we stretched the explants radially for 3 hr in a stepwise fashion (Figure 4—figure supplement 2e,f). This stepwise stretch created a force of 11.5 N and about 50–75% strain. We observed a significant increase in the apical area of MCCs (median ± SD, 409 ± 57 μm2 compared to 210 ± 43 μm2 in unstretched tethered caps) (Figure 4g–i). Stretching also led to a dramatic change in cell shape and a further significant reduction in the TR (apical area: 351–600 μm2, median ± SD, TR: 0.71 ± 0.1, Figure 4c,d) compared to both tethered unstretched caps (median ± SD, TR: 0.79 ± 0.6) and WT X. tropicalis MCCs (median ± SD, TR: 0.79 ± 0.6), consistent with the expectation that external stretching will lead to more polygonality of apical shape. In these stretched MCCs, the number of centrioles also increased (median ± SD, 199 ± 33 vs 130 ± 25 in unstretched tethered caps) demonstrating that stretching forces trigger centriole amplification in an area dependent manner in MCCs (Figure 4j). Interestingly, just by stretching, we transformed MCCs in X. tropicalis explants to sizes more similar to X. laevis (median ± SD, 409 ± 57 μm2 vs. 390 ± 87 μm2 in X. laevis) and the number of centrioles generated were also similar (median ± SD, 199 ± 33 vs. 195 ± 27 in X. laevis), highlighting the conserved role of mechanical forces in establishing the scaling mechanisms across species (Figure 1g–j, Figure 4k). Given the central role stretching plays in regulating centriole number, we decided to investigate the molecular mechanisms that sense the force. While there are several molecules that can act as mechanosensors (Luo et al., 2013; Martino et al., 2018; Wang, 2017), we were particularly struck by the punctate distribution pattern of Piezo1 (Figure 5—figure supplement 1a). Piezo1 is a mechanosensitive cation channel that responds directly to membrane stretch and is primarily expressed in epithelial cells exposed to fluid pressure and flow (Bagriantsev et al., 2014; Wang and Xiao, 2018; Wu et al., 2017). In addition to its expression at cell junctions (Figure 5—figure supplement 1b – dashed box), we unexpectedly discovered that Piezo1 is localized adjacent to the centrioles at the apical membrane (Figure 5a, Figure 5—figure supplement 1a,b). Piezo1 localization is diminished with MO-based Piezo1 depletion indicating that this anti-Piezo1 antibody signal is specific (Figure 5—figure supplement 1c–e). Figure 5 with 1 supplement see all Download asset Open asset Piezo1 fine tunes centriole amplification and the scaling relation with apical area in the embryos. (a) Mature epidermal MCCs marked with anti-Piezo1 antibody (magenta) and chibby-GFP (centrioles, green) in X. tropicalis embryos. XZ axis shows that Piezo1 localizes at the same plane as centrioles. Quantitation of (b) centriole number and (c) apical area in MCCs across different treatments that affect Piezo1 levels (MO or CRISPR) or function (GSMTx4). Dashed lines indicate the median values of controls. * indicates statistical significance at p < 0.05. The statistical comparison between the treatments is done using the Brown-Forsythe and Welch ANOVA test followed by Dunnett's T3 multiple comparisons test. n = number of cells collected from 12 to 20 embryos. MCCs marked with chibby-GFP (centrioles), and phalloidin (F-actin) in (d) Standard control MO, (e) piezo1 MO, (f) piezo1 CRISPR, and (g) GSMTx4. Regression plot demonstrat

Post-translational modification of tubulin provides differential functions to microtubule networks. Here, we address the role of tubulin acetylation on the penetrative capacity of cells undergoing radial intercalation, which is the process by which cells move apically, insert between outer cells, and join an epithelium. There are opposing forces that regulate intercalation, namely, the restrictive forces of the epithelial barrier versus the penetrative forces of the intercalating cell. Positively and negatively modulating tubulin acetylation in intercalating cells alters the developmental timing such that cells with more acetylation penetrate faster. We find that intercalating cells preferentially penetrate higher-order vertices rather than the more prevalent tricellular vertices. Differential timing in the ability of cells to penetrate different vertices reveals that lower-order vertices represent more restrictive sites of insertion. We shift the accessibility of intercalating cells toward more restrictive junctions by increasing tubulin acetylation, and we provide a geometric-based mathematical model that describes our results.

How cells count and regulate organelle number is a fundamental question in cell biology. For example, most cells restrict centrioles to two in number and assemble one cilium; however, multiciliated cells (MCCs) synthesize hundreds of centrioles to assemble multiple cilia. Aberration in centriole/cilia number impairs MCC function and can lead to pathological outcomes. Yet how MCCs control centriole number remains unknown. Using Xenopus, we demonstrate that centriole number scales with apical area over a remarkable 40-fold change in size. We find that tensile forces that shape the apical area also trigger centriole amplification based on both cell stretching experiments and disruption of embryonic elongation. Unexpectedly, Piezo1, a mechanosensitive ion channel, localizes near each centriole suggesting a potential role in centriole amplification. Indeed, depletion of Piezo1 affects centriole amplification and disrupts its correlation with the apical area in a tension-dependent manner. Thus, mechanical forces calibrate cilia/centriole number to the MCC apical area via Piezo1. Our results provide new perspectives to study organelle number control essential for optimal cell function.

Synchronized beating of cilia on multiciliated cells (MCCs) generates a directional flow of mucus across epithelia. This motility requires a "9 + 2" microtubule (MT) configuration in axonemes and the unidirectional array of basal bodies of cilia on the MCCs. However, it is not fully understood what components are needed for central MT-pair assembly as they are not continuous with basal bodies in contrast to the nine outer MT doublets. In this study, we discovered that a homozygous knockdown mouse model for MT minus-end regulator calmodulin-regulated spectrin-associated protein 3 (CAMSAP3), Camsap3(tm1a/tm1a) , exhibited multiple phenotypes, some of which are typical of primary ciliary dyskinesia (PCD), a condition caused by motile cilia defects. Anatomical examination of Camsap3(tm1a/tm1a) mice revealed severe nasal airway blockage and abnormal ciliary morphologies in nasal MCCs. MCCs from different tissues exhibited defective synchronized beating and ineffective generation of directional flow likely underlying the PCD-like phenotypes. In normal mice, CAMSAP3 localized to the base of axonemes and at the basal bodies in MCCs. However, in Camsap3(tm1a/tm1a) , MCCs lacked CAMSAP3 at the ciliary base. Importantly, the central MT pairs were missing in the majority of cilia, and the polarity of the basal bodies was disorganized. These phenotypes were further confirmed in MCCs of Xenopus embryos when CAMSAP3 expression was knocked down by morpholino injection. Taken together, we identified CAMSAP3 as being important for the formation of central MT pairs, proper orientation of basal bodies, and synchronized beating of motile cilia.

Centrioles are microtubule (MT)-based structures that provide important functions during cell migration, cell division, and cell signaling [1]. Modulating centriole number in 3D cell cultures has been shown to influence protrusive behavior [265]. Here, we address in vivo the role of centrioles and the accumulation of MTs on the protrusive behavior required during the initiation of radial intercalation. Radial intercalation is an important developmental process whereby cells undergo polarized movements and interdigitate into a more superficial layer [6, 7]. It is commonly employed during metamorphic events, such as the tissue thinning coupled with expansion or during the introduction of different cell types into an epithelium. During radial intercalation, cells emerge from a basal layer by undergoing a process of apical migration, apical insertion, and expansion [8]. In Xenopus skin, multiciliated cells (MCCs), which contain similar to 150 centrioles, and ionocytes (ICs), which contain two centrioles, differentiate during the same developmental window, but MCCs complete intercalation prior to ICs. Here, we utilize this difference in timing to create a quantifiable assay for insertion and find that the timing of insertion is modulated by changes in centriole number and the accumulation of acetylated MTs. Additionally, centrioles align between the nucleus and the leading edge creating an axis of migration with apically oriented (+) ends. Using the MT (-) end protein CAMSAP1 fused to the apically positioned Par6 protein, we have artificially reversed the orientation of MTs and find that the accumulation of MTs in either orientation is sufficient to promote apical insertion.

The basolateral protein Scribble (Scrib), a member of the LAP protein family, is essential for epithelial apicobasal polarity (ABP) in Drosophila. However, a conserved function for this protein in mammals is unclear. Here we show that the crucial role for Scrib in ABP has remained obscure due to the compensatory function of two other LAP proteins, Erbin and Lano. A combined Scrib/Erbin/Lano knockout disorganizes the cell-cell junctions and the cytoskeleton. It also results in mislocalization of several apical (Par6, aPKC, and Pals1) and basolateral (Llgl1 and Llgl2) identity proteins. These defects can be rescued by the conserved "LU" region of these LAP proteins. Structure-function analysis of this region determined that the so-called LAPSDb domain is essential for basolateral targeting of these proteins, while the LAPSDa domain is essential for supporting the membrane basolateral identity and binding to Llgl. In contrast to the key role in Drosophila, mislocalization of Llgl proteins does not appear to be critical in the scrib ABP phenotype.

Primary ciliary dyskinesia (PCD) is a genetic disorder in which impaired ciliary function leads to chronic airway disease. Exome sequencing of a PCD subject identified an apparent homozygous frameshift variant, c.887_890delTAAG (p.Val296Glyfs*13), in exon 5; this frameshift introduces a stop codon in amino acid 308 of the growth arrest-specific protein 2-like 2 (GAS2L2). Further genetic screening of unrelated PCD subjects identified a second proband with a compound heterozygous variant carrying the identical frameshift variant and a large deletion (c.867_*343+1207del; p.?) starting in exon 5. Both individuals had clinical features of PCD but normal ciliary axoneme structure. In this research, using human nasal cells, mouse models, and X. laevis embryos, we show that GAS2L2 is abundant at the apical surface of ciliated cells, where it localizes with basal bodies, basal feet, rootlets, and actin filaments. Cultured GAS2L2-deficient nasal epithelial cells from one of the affected individuals showed defects in ciliary orientation and had an asynchronous and hyperkinetic (GAS2L2-deficient = 19.8 Hz versus control = 15.8 Hz) ciliary-beat pattern. These results were recapitulated in Gas2l2(-/-) mouse tracheal epithelial cell (mTEC) cultures and in X. laevis embryos treated with Gas2l2 morpholinos. In mice, the absence of Gas2l2 caused neonatal death, and the conditional deletion of Gas2l2 impaired mucociliary clearance (MCC) and led to mucus accumulation. These results show that a pathogenic variant in GAS2L2 causes a genetic defect in ciliary orientation and impairs MCC and results in PCD.

Woolly mammoths were among the most abundant cold adapted species during the Pleistocene. Their once large populations went extinct in two waves, an end-Pleistocene extinction of continental populations followed by the mid-Holocene extinction of relict populations on St. Paul Island ~5,600 years ago and Wrangel Island ~4,000 years ago. Wrangel Island mammoths experienced an episode of rapid demographic decline coincident with their isolation, leading to a small population, reduced genetic diversity, and the fixation of putatively deleterious alleles, but the functional consequences of these processes are unclear. Here we show that the Wrangel Island mammoth accumulated many putative deleterious mutations that are predicted to cause diverse behavioral and developmental defects. Resurrection and functional characterization of Wrangel Island mammoth genes carrying these substitutions identified both loss and gain of function mutations in genes associated with developmental defects (HYLS1), oligozoospermia and reduced male fertility (NKD1), diabetes (NEUROG3), and the ability to detect floral scents (OR5A1). These results suggest that Wrangel Island mammoths may have suffered adverse consequences from their reduced population sizes and isolation.

Multiciliated cells (MCCs) amplify large numbers of centrioles that convert into basal bodies, which are required for producing multiple motile cilia. Most centrioles amplified by MCCs grow on the surface of organelles called deuterosomes, whereas a smaller number grow through the centriolar pathway in association with the two parent centrioles. Here, we show that MCCs lacking deuterosomes amplify the correct number of centrioles with normal step-wise kinetics. This is achieved through a massive production of centrioles on the surface and in the vicinity of parent centrioles. Therefore, deuterosomes may have evolved to relieve, rather than supplement, the centriolar pathway during multiciliogenesis. Remarkably, MCCs lacking parent centrioles and deuterosomes also amplify the appropriate number of centrioles inside a cloud of pericentriolar and fibrogranular material. These data show that the centriole number is set independently of their nucleation platforms and suggest that massive centriole production in MCCs is a robust process that can self-organize.

Cilia are microtubule-based structures that protrude from the cell surface. Primary cilia function as sensors for environmental signals to regulate cell fate decisions and motile cilia produce directed fluid flow. In multiciliated cells (MCCs) there is a dynamic and intricate process that allows for the generation of greater than 150 evenly-spaced cilia, which coordinate to produce directed fluid flow over the epithelial surface. Induction of this MCC fate is restricted by Notch signaling, whereas activation of Notch signaling promotes secretory cell formation. It has been proposed that MCCs have the potential to resorb their cilia and transdifferentiate into secretory cells, however the mechanisms coordinating this transdifferentiation processes remain unknown. In order to address this, we utilized Xenopus embryonic skin as a model system. In Xenopus skin development, MCCs intercalate into to the outer epithelial layer that is composed of secretory cells. We have found that as Xenopus embryos continue to mature MCCs begin to lose their cilia-driven fluid flow as a result of cilia resorption followed by transdifferentiation into secretory cells. At this time, MCCs generate large intracellular vacuoles and the autophagosome protein LC3 localizes to the base of cilia suggesting that autophagy of cilia components, ciliophagy, is an important process in transdifferentiation. Since MCCs are a fully differentiated cell, the potential to coordinate large-scale ciliophagy for transdifferentiation represents a surprising and remarkable process. Our results demonstrate that Xenopus skin can be used to better understand the mechanistic links between cilia resorption, autophagy, and Notch signaling during transdifferentiation of epithelial cells.