A rapid segmentation method of cell boundary for developing embryos using machine learning with a personal computer.

Cell segmentation is crucial in the study of morphogenesis in developing embryos, but it had been limited in its accuracy until machine learning methods for image segmentation like U-Net. However, these methods take too much time. In this study, we provide a rapid method for cell segmentation using machine learning with a personal computer, termed Cell Segmentator using Machine Learning (CSML). CSML took four seconds per image with a personal computer for segmentation on average, much less than time to obtain an image. We observed that F-value of segmentation by CSML was around 0.97, showing better performance than state-of-the-art methods like RACE and watershed in assessing the segmentation of Xenopus ectodermal cells. CSML also showed slightly better performance and faster than other machine learning-based methods such as U-Net. CSML required only one whole embryo image for training a Fully Convolutional Network classifier and only two parameters. To validate its accuracy, we compared CSML to other methods in assessing several indicators of cell shape. We also examined the generality of this approach by measuring its performance of segmentation of independent images. Our data demonstrate the superiority of CSML, and we expect this application to improve efficiency in cell shape studies.

Tracheal motile cilia in mice require CAMSAP3 for the formation of central microtubule pair and coordinated beating.

Motile cilia of multiciliated epithelial cells undergo synchronized beating to produce fluid flow along the luminal surface of various organs. Each motile cilium consists of an axoneme and a basal body (BB), which are linked by a "transition zone" (TZ). The axoneme exhibits a characteristic 9+2 microtubule arrangement important for ciliary motion, but how this microtubule system is generated is not yet fully understood. Here we show that calmodulin-regulated spectrin-associated protein 3 (CAMSAP3), a protein that can stabilize the minus-end of a microtubule, concentrates at multiple sites of the cilium-BB complex, including the upper region of the TZ or the axonemal basal plate (BP) where the central pair of microtubules (CP) initiates. CAMSAP3 dysfunction resulted in loss of the CP and partial distortion of the BP, as well as the failure of multicilia to undergo synchronized beating. These findings suggest that CAMSAP3 plays pivotal roles in the formation or stabilization of the CP by localizing at the basal region of the axoneme and thereby supports the coordinated motion of multicilia in airway epithelial cells.

CFAP53 regulates mammalian cilia-type motility patterns through differential localization and recruitment of axonemal dynein components.

Motile cilia can beat with distinct patterns, but how motility variations are regulated remain obscure. Here, we have studied the role of the coiled-coil protein CFAP53 in the motility of different cilia-types in the mouse. While node (9+0) cilia of Cfap53 mutants were immotile, tracheal and ependymal (9+2) cilia retained motility, albeit with an altered beat pattern. In node cilia, CFAP53 mainly localized at the base (centriolar satellites), whereas it was also present along the entire axoneme in tracheal cilia. CFAP53 associated tightly with microtubules and interacted with axonemal dyneins and TTC25, a dynein docking complex component. TTC25 and outer dynein arms (ODAs) were lost from node cilia, but were largely maintained in tracheal cilia of Cfap53-/- mice. Thus, CFAP53 at the base of node cilia facilitates axonemal transport of TTC25 and dyneins, while axonemal CFAP53 in 9+2 cilia stabilizes dynein binding to microtubules. Our study establishes how differential localization and function of CFAP53 contributes to the unique motion patterns of two important mammalian cilia-types.

CFAP45 deficiency causes situs abnormalities and asthenospermia by disrupting an axonemal adenine nucleotide homeostasis module.

Axonemal dynein ATPases direct ciliary and flagellar beating via adenosine triphosphate (ATP) hydrolysis. The modulatory effect of adenosine monophosphate (AMP) and adenosine diphosphate (ADP) on flagellar beating is not fully understood. Here, we describe a deficiency of cilia and flagella associated protein 45 (CFAP45) in humans and mice that presents a motile ciliopathy featuring situs inversus totalis and asthenospermia. CFAP45-deficient cilia and flagella show normal morphology and axonemal ultrastructure. Proteomic profiling links CFAP45 to an axonemal module including dynein ATPases and adenylate kinase as well as CFAP52, whose mutations cause a similar ciliopathy. CFAP45 binds AMP in vitro, consistent with structural modelling that identifies an AMP-binding interface between CFAP45 and AK8. Microtubule sliding of dyskinetic sperm from Cfap45-/- mice is rescued with the addition of either AMP or ADP with ATP, compared to ATP alone. We propose that CFAP45 supports mammalian ciliary and flagellar beating via an adenine nucleotide homeostasis module.

Role of Ca2+ transients at the node of the mouse embryo in breaking of left-right symmetry.

Immotile cilia sense extracellular signals such as fluid flow, but whether Ca2+ plays a role in flow sensing has been unclear. Here, we examined the role of ciliary Ca2+ in the flow sensing that initiates the breaking of left-right (L-R) symmetry in the mouse embryo. Intraciliary and cytoplasmic Ca2+ transients were detected in the crown cells at the node. These Ca2+ transients showed L-R asymmetry, which was lost in the absence of fluid flow or the PKD2 channel. Further characterization allowed classification of the Ca2+ transients into two types: cilium-derived, L-R-asymmetric transients (type 1) and cilium-independent transients without an L-R bias (type 2). Type 1 intraciliary transients occurred preferentially at the left posterior region of the node, where L-R symmetry breaking takes place. Suppression of intraciliary Ca2+ transients delayed L-R symmetry breaking. Our results implicate cilium-derived Ca2+ transients in crown cells in initiation of L-R symmetry breaking in the mouse embryo.

Nodal paralogues underlie distinct mechanisms for visceral left-right asymmetry in reptiles and mammals.

Unidirectional fluid flow generated by motile cilia at the left-right organizer (LRO) breaks left-right (L-R) symmetry during early embryogenesis in mouse, frog and zebrafish. The chick embryo, however, does not require motile cilia for L-R symmetry breaking. The diversity of mechanisms for L-R symmetry breaking among vertebrates and the trigger for such symmetry breaking in non-mammalian amniotes have remained unknown. Here we examined how L-R asymmetry is established in two reptiles, Madagascar ground gecko and Chinese softshell turtle. Both of these reptiles appear to lack motile cilia at the LRO. The expression of the Nodal gene at the LRO in the reptilian embryos was found to be asymmetric, in contrast to that in vertebrates such as mouse that are dependent on cilia for L-R patterning. Two paralogues of the Nodal gene derived from an ancient gene duplication are retained and expressed differentially in cilia-dependent and cilia-independent vertebrates. The expression of these two Nodal paralogues is similarly controlled in the lateral plate mesoderm but regulated differently at the LRO. Our in-depth analysis of reptilian embryos thus suggests that mammals and non-mammalian amniotes deploy distinct strategies dependent on different Nodal paralogues for rendering Nodal activity asymmetric at the LRO.

Loss of Fam60a, a Sin3a subunit, results in embryonic lethality and is associated with aberrant methylation at a subset of gene promoters.

We have examined the role of Fam60a, a gene highly expressed in embryonic stem cells, in mouse development. Fam60a interacts with components of the Sin3a-Hdac transcriptional corepressor complex, and most Fam60a-/- embryos manifest hypoplasia of visceral organs and die in utero. Fam60a is recruited to the promoter regions of a subset of genes, with the expression of these genes being either up- or down-regulated in Fam60a-/- embryos. The DNA methylation level of the Fam60a target gene Adhfe1 is maintained at embryonic day (E) 7.5 but markedly reduced at E9.5 in Fam60a-/- embryos, suggesting that DNA demethylation is enhanced in the mutant. Examination of genome-wide DNA methylation identified several differentially methylated regions, which were preferentially hypomethylated, in Fam60a-/- embryos. Our data suggest that Fam60a is required for proper embryogenesis, at least in part as a result of its regulation of DNA methylation at specific gene promoters.

Chlamydomonas DYX1C1/PF23 is essential for axonemal assembly and proper morphology of inner dynein arms.

Cytoplasmic assembly of ciliary dyneins, a process known as preassembly, requires numerous non-dynein proteins, but the identities and functions of these proteins are not fully elucidated. Here, we show that the classical Chlamydomonas motility mutant pf23 is defective in the Chlamydomonas homolog of DYX1C1. The pf23 mutant has a 494 bp deletion in the DYX1C1 gene and expresses a shorter DYX1C1 protein in the cytoplasm. Structural analyses, using cryo-ET, reveal that pf23 axonemes lack most of the inner dynein arms. Spectral counting confirms that DYX1C1 is essential for the assembly of the majority of ciliary inner dynein arms (IDA) as well as a fraction of the outer dynein arms (ODA). A C-terminal truncation of DYX1C1 shows a reduction in a subset of these ciliary IDAs. Sucrose gradients of cytoplasmic extracts show that preassembled ciliary dyneins are reduced compared to wild-type, which suggests an important role in dynein complex stability. The role of PF23/DYX1C1 remains unknown, but we suggest that DYX1C1 could provide a scaffold for macromolecular assembly.

Eyespot-dependent determination of the phototactic sign in Chlamydomonas reinhardtii.

The biflagellate green alga Chlamydomonas reinhardtii exhibits both positive and negative phototaxis to inhabit areas with proper light conditions. It has been shown that treatment of cells with reactive oxygen species (ROS) reagents biases the phototactic sign to positive, whereas that with ROS scavengers biases it to negative. Taking advantage of this property, we isolated a mutant, lts1-211, which displays a reduction-oxidation (redox) dependent phototactic sign opposite to that of the wild type. This mutant has a single amino acid substitution in phytoene synthase, an enzyme that functions in the carotenoid-biosynthesis pathway. The eyespot contains large amounts of carotenoids and is crucial for phototaxis. Most lts1-211 cells have no detectable eyespot and reduced carotenoid levels. Interestingly, the reversed phototactic-sign phenotype of lts1-211 is shared by other eyespot-less mutants. In addition, we directly showed that the cell body acts as a convex lens. The lens effect of the cell body condenses the light coming from the rear onto the photoreceptor in the absence of carotenoid layers, which can account for the reversed-phototactic-sign phenotype of the mutants. These results suggest that light-shielding property of the eyespot is essential for determination of phototactic sign.

Identification of the agg1 mutation responsible for negative phototaxis in a "wild-type" strain of Chlamydomonas reinhardtii.

The unicellular green alga Chlamydomonas reinhardtii is a model organism for various studies in biology. CC-124 is a laboratory strain widely used as a wild type. However, this strain is known to carry agg1 mutation, which causes cells to swim away from the light source (negative phototaxis), in contrast to the cells of other wild-type strains, which swim toward the light source (positive phototaxis). Here we identified the causative gene of agg1 (AGG1) using AFLP-based gene mapping and whole genome next-generation sequencing. This gene encodes a 36-kDa protein containing a Fibronectin type III domain and a CHORD-Sgt1 (CS) domain. The gene product is localized to the cell body and not to flagella or basal body.

TTC26/DYF13 is an intraflagellar transport protein required for transport of motility-related proteins into flagella.

Cilia/flagella are assembled and maintained by the process of intraflagellar transport (IFT), a highly conserved mechanism involving more than 20 IFT proteins. However, the functions of individual IFT proteins are mostly unclear. To help address this issue, we focused on a putative IFT protein TTC26/DYF13. Using live imaging and biochemical approaches we show that TTC26/DYF13 is an IFT complex B protein in mammalian cells and Chlamydomonas reinhardtii. Knockdown of TTC26/DYF13 in zebrafish embryos or mutation of TTC26/DYF13 in C. reinhardtii, produced short cilia with abnormal motility. Surprisingly, IFT particle assembly and speed were normal in dyf13 mutant flagella, unlike in other IFT complex B mutants. Proteomic and biochemical analyses indicated a particular set of proteins involved in motility was specifically depleted in the dyf13 mutant. These results support the concept that different IFT proteins are responsible for different cargo subsets, providing a possible explanation for the complexity of the IFT machinery. DOI: http://dx.doi.org/10.7554/eLife.01566.001.

Protein-protein interactions between intermediate chains and the docking complex of Chlamydomonas flagellar outer arm dynein.

Outer arm dynein (OAD) is bound to specific loci on outer-doublet-microtubules by interactions at two sites: via intermediate chain 1 (IC1) and the outer dynein arm docking complex (ODA-DC). Studies using Chlamydomonas mutants have suggested that the individual sites have rather weak affinities for microtubules, and therefore strong OAD attachment to microtubules is achieved by their cooperation. To test this idea, we examined interactions between IC1, IC2 (another intermediate chain) and ODA-DC using recombinant proteins. Recombinant IC1 and IC2 were found to form a 1:1 complex, and this complex associated with ODA-DC in vitro. Binding of IC1 to mutant axonemes revealed that there are specific binding sites for IC1. From these data, we propose a novel model of OAD-outer doublet association.

Herpes simplex virus 1 VP22 regulates translocation of multiple viral and cellular proteins and promotes neurovirulence.

Herpes simplex virus 1 (HSV-1) protein VP22, encoded by the UL49 gene, is a major virion tegument protein. In the present study, we showed that VP22 was required for efficient redistribution of viral proteins VP16, VP26, ICP0, ICP4, and ICP27 and of cellular protein Hsc-70 to the cytoplasm of infected cells. We found that two dileucine motifs in VP22, at amino acids 235 and 236 and amino acids 251 and 252, were necessary for VP22 regulation of the proper cytoplasmic localization of these viral and cellular proteins. The dileucine motifs were also required for proper cytoplasmic localization of VP22 itself and for optimal expression of viral proteins VP16, VP22, ICP0, UL41, and glycoprotein B. Interestingly, a recombinant mutant virus with alanines substituted for the dileucines at amino acids 251 and 252 had a 50% lethal dose (LD(50)) for neurovirulence in mice following intracerebral inoculation about 10(3)-fold lower than the LD(50) of the repaired virus. Furthermore, the replication and spread of this mutant virus in the brains of mice following intracerebral inoculation were significantly impaired relative to those of the repaired virus. The ability of VP22 to regulate the localization and expression of various viral and cellular proteins, as shown in this study, was correlated with an increase in viral replication and neurovirulence in the experimental murine model. Thus, HSV-1 VP22 is a significant neurovirulence factor in vivo.

Calcium-dependent flagellar motility activation in Chlamydomonas reinhardtii in response to mechanical agitation.

Flagellar beating in Chlamydomonas was found to be activated by mechanical stimulation. Immediately after a wild-type cell suspension was vortexed, the average swimming velocity of cells increased from 130 mum/second to 150 mum/second, due to an elevation of flagellar beat frequency from approximately 60 Hz to approximately 70 Hz without detectable change in the flagellar waveforms. This response required outer arm dynein. Treatment with EGTA, Ca(2+)-channel blockers, or mechanosensitive-channel blockers inhibited it. In demembranated and reactivated cell models, a modest increase in Ca(2+) concentration elevated the axonemal beat frequency. These data indicate that the mechanical agitation increases beat frequency because it causes Ca(2+) influx into flagella, which then activates outer arm dynein.