Microglia are specialized macrophages responsible for the clearance of dead neurons and pathogens by phagocytosis and degradation. The degradation requires phagosome maturation and acidification provided by the vesicular- or vacuolar-type H+-translocating adenosine triphosphatase (V-ATPase), which is composed of the cytoplasmic V1 domain and the membrane-embedded Vo domain. The V-ATPase a subunit, an integral part of the Vo domain, has four isoforms in mammals. The functions of different isoforms on phagosome maturation in different cells/species remain controversial. Here we show that mutations of both the V-ATPase Atp6v0a1 and Tcirg1b/Atp6v0a3 subunits lead to the accumulation of phagosomes in zebrafish microglia. However, their mechanisms are different. The V-ATPase Atp6v0a1 subunit is mainly distributed in early and late phagosomes. Defects of this subunit lead to a defective transition from early phagosomes to late phagosomes. In contrast, The V-ATPase Tcirg1b/Atp6v0a3 subunit is primarily located on lysosomes and regulates late phagosome-lysosomal fusion. Defective Tcirg1b/Atp6v0a3, but not Atp6v0a1 subunit leads to reduced acidification and impaired macroautophagy/autophagy in microglia. We further showed that ATP6V0A1/a1 and TCIRG1/a3 subunits in mouse macrophages preferentially located in endosomes and lysosomes, respectively. Blocking these subunits disrupted early-to-late endosome transition and endosome-to-lysosome fusion, respectively. Taken together, our results highlight the essential and conserved roles played by different V-ATPase subunits in multiple steps of phagocytosis and endocytosis across various species.Abbrevations: Apoe: apolipoprotein E; ANXA5/annexin V: annexin A5; ATP6V0A1/a1: ATPase H+-transporting V0 subunit a1; ATP6V0A2/a2: ATPase H+-transporting V0 subunit a2; ATP6V0A4/a4: ATPase H+-transporting V0 subunit a4; dpf: days post-fertilization; EEA1: early endosome antigen 1; HOPS: homotypic fusion and protein sorting; LAMP1: lysosomal associated membrane protein 1; Lcp1: lymphocyte cytosolic protein 1 (L-plastin); Map1lc3/Lc3: microtubule-associated protein 1 light chain 3; NR: neutral red; PBS: phosphate-buffered saline; PtdIns: phosphatidylinositol; PtdIns3P: phosphatidylinositol-3-phosphate; PtdIns(3,5)P2: phosphatidylinositol (3,5)-bisphosphate; RAB4: RAB4, member RAS oncogene family; RAB5: RAB5, member RAS oncogene family; RAB7: RAB7, member RAS oncogene family; TCIRG1/Atp6v0a3/a3: T cell immune regulator 1, ATPase H+-transporting V0 subunit a3; V-ATPase: vacuolar-type H+-translocating adenosine triphosphatase; Xla.Tubb2b/NBT: tubulin beta 2B class IIb.

Mesenchymal stromal cells are essential components of hematopoietic stem and progenitor cell (HSPC) niches, regulating HSPC proliferation and fates. Their developmental origins are largely unknown. In zebrafish, we previously found that the stromal cells of the caudal hematopoietic tissue (CHT), a niche functionally homologous to the mammalian fetal liver, arise from the ventral part of caudal somites. We have now found that this ventral domain is the sclerotome, and that two markers of mammalian mesenchymal stem/stromal cells, Alcam and Pdgfr-α, are distinctively expressed there and instrumental for the emergence and migration of stromal cell progenitors, which in turn conditions the proper assembly of the vascular component of the CHT niche. Furthermore, we find that trunk somites are similarly dependent on Alcam and Pdgfr-α to produce mesenchymal cells that foster HSPC emergence from the aorta. Thus the sclerotome contributes essential stromal cells for each of the key steps of developmental hematopoiesis.

Trim33 (Tif1γ) is a transcriptional regulator that is notably involved in several aspects of hematopoiesis. It is essential for the production of erythrocytes in zebrafish, and for the proper functioning and aging of hematopoietic stem and progenitor cells (HSPCs) in mice. Here, we have found that, in zebrafish development, Trim33 is essential cell-autonomously for the lifespan of the yolk sac-derived primitive macrophages, as well as for the initial production of definitive (HSPC-derived) macrophages in the first niche of definitive hematopoiesis, the caudal hematopoietic tissue. Moreover, Trim33 deficiency leads to an excess production of definitive neutrophils and thrombocytes. Our data indicate that Trim33 radically conditions the differentiation output of aorta-derived HSPCs in all four erythro-myeloid cell types, in a niche-specific manner.

Hematopoietic stem and progenitor cells emerge from the aorta and migrate to the caudal hematopoietic tissue (CHT) of zebrafish larvae, the hematopoietic equivalent of the mammalian fetal liver, for their proliferation and differentiation. We previously reported that somite-derived stromal cells were a key component of the CHT niche. Here, we found that the cell adhesion protein Protocadherin 18a (Pcdh18a) is expressed in the stromal cell progenitors (SCPs) emigrating from somites toward the future CHT. Deletion of most of the Pcdh18a intracellular domain caused a decrease in the number of SCPs, the directionality of their migration, and the cell-contact mediated repulsion that normally occurs between migrating SCPs. These defects were followed by abnormal morphogenesis of the venous plexus that forms the CHT framework, and the inability of the CHT to function as a niche for hematopoietic stem and progenitor cells. Finally, we found that the extracellular domain of Pcdh18a mediates trans heterophilic adhesion of stromal cells to endothelial cells in vivo and thereby the reticular versus perivascular fate of SCPs. Thus, Pcdh18a expression in SCPs is essential for the proper development of the hematopoietic niche.

Hematopoietic stem and progenitor cells (HSPCs) are multipotent cells giving rise to all blood lineages during life. HSPCs emerge from the ventral wall of the dorsal aorta (VDA) during a specific timespan in embryonic development through endothelial hematopoietic transition (EHT). We investigated the ontogeny of HSPCs in mutant zebrafish embryos lacking functional pten, an important tumor suppressor with a central role in cell signaling. Through in vivo live imaging, we discovered that in pten mutant embryos a proportion of the HSPCs died upon emergence from the VDA, an effect rescued by inhibition of phosphatidylinositol-3 kinase (PI3K). Surprisingly, inhibition of PI3K in wild-type embryos also induced HSPC death. Surviving HSPCs colonized the caudal hematopoietic tissue (CHT) normally and committed to all blood lineages. Single-cell RNA sequencing indicated that inhibition of PI3K enhanced survival of multipotent progenitors, whereas the number of HSPCs with more stem-like properties was reduced. At the end of the definitive wave, loss of Pten caused a shift to more restricted progenitors at the expense of HSPCs. We conclude that PI3K signaling tightly controls HSPCs survival and both up- and downregulation of PI3K signaling reduces stemness of HSPCs.

Hematopoietic Stem and Progenitor Cells (HSPCs) are multipotent cells giving rise to all blood lineages during life. HSPCs emerge from the ventral wall of the dorsal aorta (VDA) during a specific timespan in embryonic development through endothelial hematopoietic transition (EHT). We investigated the ontogeny of HSPCs in mutant zebrafish embryos lacking functional pten, an important tumor suppressor with a central role in cell signaling. Through in vivo live imaging, we discovered that in pten mutant embryos a proportion of the HSPCs died upon emergence from the VDA, an effect rescued by inhibition of phosphatidylinositol-3 kinase (PI3K). Surprisingly, inhibition of PI3K in wild type embryos also induced HSPC death. Surviving HSPCs colonized the caudal hematopoietic tissue (CHT) normally and committed to all blood lineages. Single cell RNA sequencing indicated that inhibition of PI3K enhanced survival of multi-potent progenitors, whereas the number of HSPCs with more stem-like properties was reduced. At the end of the definitive wave, loss of Pten caused a shift to more restricted progenitors at the expense of HSPCs. We conclude that PI3K signaling tightly controls HSPCs survival and both up- and downregulation of PI3K signaling reduces stemness of HSPCs. 2 Key points Loss of Pten and inhibition of PI3K induced apoptosis of hematopoietic stem/progenitor cells upon endothelial to hematopoietic transition Surviving hematopoietic stem/progenitor cells committed to all blood lineages but displayed reduced stemness

Most tissues harbor a substantial population of resident macrophages. Here, we elucidate a functional link between the Slc7a7 cationic amino acid transporter and tissue macrophages. We identified a mutant zebrafish devoid of microglia due to a mutation in the slc7a7 gene. We found that in Slc7a7-deficient larvae, macrophages do enter the retina and brain to become microglia, but then die during the developmental wave of neuronal apoptosis, which triggers intense efferocytic work from them. A similar macrophage demise occurs in other tissues, at stages where macrophages have to engulf many cell corpses, whether due to developmental or experimentally triggered cell death. We found that Slc7a7 is the main cationic amino acid transporter expressed in macrophages of zebrafish larvae, and that its expression is induced in tissue macrophages within 1-2 h upon efferocytosis. Our data indicate that Slc7a7 is vital not only for microglia but also for any steadily efferocytic tissue macrophages, and that slc7a7 gene induction is one of the adaptive responses that allow them to cope with the catabolism of numerous dead cells without compromising their own viability.

Most tissues harbor a substantial population of resident macrophages. It is not quite known yet how their quite diverse phenotypes are shaped by the functions that they assume in each tissue. In this study, we elucidate a functional link between the Slc7a7 cationic amino acid transporter and tissue macrophages. We had identified a mutant zebrafish devoid of microglia due to a mutation in the slc7a7 gene. We found that in Slc7a7 deficient larvae, macrophages do enter the retina and brain to become microglia, but then die during the developmental wave of neuronal apoptosis, which triggers intense efferocytic work from them. A similar macrophage demise occurs at other tissues and stages whereby macrophages have to engulf many cell corpses, be it due to developmental or experimentally triggered cell death. We found that slc7a7 is by far the main cationic amino acid transporter gene expressed in macrophages of wild type zebrafish larvae, and that its expression is induced in tissue macrophages within 1-2 hrs upon efferocytosis. Our data altogether indicate that a high level of Slc7a7 is vital not only for microglia but also for any steadily efferocytic tissue macrophages, and that slc7a7 gene induction is one of the adaptive responses that allow them to cope with the catabolism of numerous dead cells without compromising their own viability. ### Competing Interest Statement The authors have declared no competing interest.

A fluorogenic benzochalcone specifically labels live neutrophil granules in whole wild-type, GFP- or RFP-expressing zebrafish embryos and larvae.

Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Appendix 1 Appendix 2 Appendix 3 Appendix 4 Appendix 5 Data availability References Decision letter Author response Article and author information Metrics Abstract Heterogeneity broadly exists in various cell types both during development and at homeostasis. Investigating heterogeneity is crucial for comprehensively understanding the complexity of ontogeny, dynamics, and function of specific cell types. Traditional bulk-labeling techniques are incompetent to dissect heterogeneity within cell population, while the new single-cell lineage tracing methodologies invented in the last decade can hardly achieve high-fidelity single-cell labeling and long-term in-vivo observation simultaneously. In this work, we developed a high-precision infrared laser-evoked gene operator heat-shock system, which uses laser-induced CreERT2 combined with loxP-DsRedx-loxP-GFP reporter to achieve precise single-cell labeling and tracing. In vivo study indicated that this system can precisely label single cell in brain, muscle and hematopoietic system in zebrafish embryo. Using this system, we traced the hematopoietic potential of hemogenic endothelium (HE) in the posterior blood island (PBI) of zebrafish embryo and found that HEs in the PBI are heterogeneous, which contains at least myeloid unipotent and myeloid-lymphoid bipotent subtypes. eLife digest Animals begin life as a single cell that then divides to become a complex organism with many different types of cells. Every time a cell divides, each of its two daughter cells can either stay the same type as their parent or adopt a different identity. Once a cell acquires an identity, it usually cannot ‘go back’ and choose another. Eventually, this process will produce daughter cells with the identity of a specific tissue or organ and that cannot divide further. Multipotent cells are cells that can produce daughter cells with different identities, including other multipotent cells. These cells can usually give rise to different cell types in a specific organ, and generate more cells to replace any cells that die in that organ. Tracking the cells descended from a multipotent cell in a specific tissue can provide information about how the tissue develops. Hemogenic endothelium cells produce the multipotent cells that give rise to two types of white blood cells: myeloid cells and lymphoid cells. Myeloid cells include innate immune cells that protect the body from infection non-specifically; while lymphoid cells include T cells and B cells with receptors that detect specific bacteria or viruses. It remains unclear whether each of these two cell types originate from a single population of hemogenic endothelium cells or from two distinct subpopulations. He et al. have now developed a new optical technique to label a single hemogenic endothelium cell in a zebrafish and track the cell and its descendants. This method revealed that there are at least two distinct populations of hemogenic endothelium cells. One of them can give rise to both lymphoid and myeloid cells, while the other can only give rise to myeloid cells. These findings shed light on the mechanisms of blood formation, and potentially could provide useful tools to study the development of diseases such as leukemia. Additionally, the single-cell labeling technology He et al. have developed could be applied to study the development of other tissues and organs. Introduction Since new cells are generated from pre-existing cells, the frequently asked questions are what progenies are generated from certain pre-existing cells and how they contribute to the organism (Child, 1906; Hoppe et al., 2014; Spanjaard and Junker, 2017). To address these questions, fate mapping has been widely used as a crucial methodology to identify progenies of the targeted cells and to trace their location, differentiation and functional dynamics (Kretzschmar and Watt, 2012). Hematopoiesis, the process of forming blood cells (Dharampuriya et al., 2017), is an outstanding paradigm for studying these issues in different animal models (Höfer et al., 2016; Jagannathan-Bogdan and Zon, 2013; Orkin and Zon, 2008). Zebrafish has natural advantages on hematopoietic fate mapping owing to the external development, transparent embryo body and its highly conserved hematopoiesis (Jagannathan-Bogdan and Zon, 2013; Stachura and Traver, 2016). Taking these advantages, permanent genetic marking, photo-convertible labeling and in vivo time-lapse imaging have been employed to monitor the generation, mobilization and lineage specification of hematopoietic stem/progenitor cells (HSPCs) in zebrafish (Murayama et al., 2006; Jin et al., 2007; Bertrand et al., 2010; Kissa and Herbomel, 2010). Recently, the infrared laser-evoked gene operator (IR-LEGO) microscope heating system (Deguchi, 2009; Kamei et al., 2009) has been demonstrated as a powerful tool for bulk cell tracing with high temporal-spatial resolution (Shimada et al., 2013; Okuyama et al., 2013; Xu et al., 2015; Tian et al., 2017; Singhal and Shaham, 2017; He et al., 2018; Henninger et al., 2017). In this system, an infrared (IR) laser is used to generate local heat shock to induce CreER expression in a restricted region of transgenic fish carrying a tissue-specific loxP-DsRedx-loxP-GFP reporter and a hsp70l:mCherry-T2a-CreERT2. The removal of DsRedx cassette is permanently inherited so that the progenies derived from the targeted tissue will display GFP instead of DsRedx (Xu et al., 2015; Tian et al., 2017; He et al., 2018). However, it has been realized that heterogeneity broadly exists in multiple cell populations during hematopoiesis (Tian et al., 2017; Wilson et al., 2015; Chen et al., 2011; Crisan and Dzierzak, 2016; Ye et al., 2017). The dissection of heterogeneity requires a lineage tracing strategy with single cell resolution. Yet, the previous IR-LEGO techniques (Deguchi, 2009; Kamei et al., 2009; Shimada et al., 2013; Okuyama et al., 2013; Xu et al., 2015; Tian et al., 2017; Singhal and Shaham, 2017; He et al., 2018; Kawasumi‐Kita, 2015; Suzuki et al., 2014; Eiji, 2013; Hayashi et al., 2014; Miao and Hayashi, 2015; Hasugata et al., 2018) face the following fundamental challenges: (1) high-precision and efficient labeling of the targeted single cell; (2) fine balance between labeling efficiency and cell viability after heat shock treatment; (3) permanent marking and long-term tracing of all progeny of the labeled single cell; (4) rigorous statistical analysis to quantitatively determine the lineages of the labeled single cell under random basal interference. These challenges hamper the wider application of IR-LEGO for cell fate mapping, and it remains unclear whether this technique can indeed be used for long-term tracing of multiple lineages of a single multipotent cell, such as HSPC. Besides IR-LEGO technique, other single-cell lineage tracing methodologies invented in past decade suffer from similar problems. Cell barcode techniques, either by retroviral library infection to insert inheritable DNA barcodes (Naik et al., 2014), or by CRISPR/Cas9 system to accumulate random mutations (McKenna et al., 2016; Kalhor et al., 2017), have been used to perform single-cell lineage tracing in hematopoiesis studies (Gerrits et al., 2010; Lu et al., 2011). Single-cell RNA-sequencing is also a prevalent way to depict lineage hierarchy (Hoppe et al., 2014; Zhou et al., 2016; Athanasiadis et al., 2017). However, these non-imaging based techniques are not suitable for tracing the dynamic behaviors of the targeted cells and their progenies. A multicolor strategy, which stochastically expresses multiple fluorescent reporters in target cells via Cre-mediated recombination, forming dozens of different color modes to distinguish individual cells and their progenies, has been utilized for cell fate mapping (Livet et al., 2007; Cai et al., 2013). Despite its success in some zebrafish hematopoiesis study (Henninger et al., 2017; Pan et al., 2013), this multicolor labeling of cells makes it difficult to directly visualize the development of individual cell lineage. In addition, the lineage hierarchy could be misinterpreted when unrelated cells share the same color or Cre-mediated recombination occurs in the daughter cells. Likewise, photo-convertible protein or caged fluorescent dyes approaches also have limitation for long-term tracing due to the self-degradation and rapid dilution of fluorophores during cell division (Tian et al., 2017; Warga et al., 2009). An optical uncaging method was used to label a targeted cell and its progeny in zebrafish through Cre-based gene recombination (Sinha et al., 2010a; Tekeli et al., 2016). However, the basal uncaging level of the caged compound in zebrafish embryo was as high as 28% (Sinha et al., 2010b), limiting the application of this technique to trace the long-term development of highly dynamic cells, such as stem cells. Thus, developing a highly precise single-cell labeling method for the long-term in vivo tracing of individual cells will be important for understanding the heterogeneity of HSPCs. To overcome the drawbacks of existing techniques, we develop a high-precision single-cell IR-LEGO technology, in which a two-photon fluorescent thermometer is utilized to measure the temperature rise in vivo to achieve precise single-cell labeling. Using this tool, we document that the hemogenic endothelium (HE) cells in the posterior blood island (PBI) of zebrafish are heterogeneous in terms of hematopoietic potential. Our study demonstrates that the high-precision single-cell IR-LEGO technology has outstanding capacity to perform single-cell labeling and long-term in-vivo lineage tracing. Results Single-cell IR-LEGO technology A 1,342 nm diode-pumped solid-state (DPSS) IR laser is used as the heat-shock light source in our single-cell IR-LEGO heat-shock microscope system (Figure 1A). The laser at this wavelength provides an appropriate balance between the absorption efficiency of water and penetration depth in tissue (Appendix 1). The IR laser is integrated with a two-photon microscope, and it is guided by the two-photon fluorescence imaging to heat the targeted cell. A water-immersion objective with a large numerical aperture (NA) is used to generate highly localized and stable laser heating at different depths in tissues. The numerical simulation (Figure 1—figure supplement 1 and Appendix 2) shows that heat shock of high spatial resolution can be achieved by generating a point heat source inside tissue, a medium of relatively poor thermal conductivity. A large temperature gradient can be created in the region of about 10 µm size around the heat source (Figure 1—figure supplement 1D), suggesting that the thermal energy produced by a highly focused IR laser heating in tissue could be confined within the single-cell dimensions for efficient single-cell gene induction. Figure 1 with 2 supplements see all Download asset Open asset IR-LEGO heat-shock microscopy and fluorescent thermometry. (A) Schematic diagram of the integrated heat-shock microscope system and fluorescent thermometry. A 1,342 nm continuous-wave (CW) IR laser was used for localized heat shock, and a femtosecond laser was used for TPEF imaging and spectroscopy. The femtosecond laser and CW laser beams were individually controlled by two pairs of galvanometer scanners. DM: dichroic mirror; PMT: photomultiplier tube; FM: foldable mirror; SP: short-pass filter; FB: fiber bundle; EMCCD: electron multiplying charge-coupled device; L1-L6: relay lens; SHG: second harmonic generation signals. (B, C) TPEF spectra of TAMRA and FITC at different temperatures, respectively. (D) Calibration of temperature sensitivity: fluorescence intensity ratio (TAMRA/FITC) as a function of temperature in water, 3% agarose (tissue phantom) and zebrafish muscle in vivo. (E) Representative results of measured temperature vs. actual temperature in 3% agarose. Slope of linear fitting: 0.998. Since the heat shock efficiency varies as the type and location of targeted cells, it is of great significance to develop a reliable method that can objectively determine the optimal IR laser heating conditions for single-cell gene induction. Although previous studies demonstrated that temperature-sensitive fluorescent proteins, such as GFP and mCherry, can be used as thermometers to estimate the temperature rise in cells induced by IR laser irradiation (Kamei et al., 2009; Singhal and Shaham, 2017), this single-molecular/one-color thermometry has been shown to produce significant errors, likely because of the fluctuation of excitation laser power, or the interference of complex microenvironment on signal intensities of fluorescence emission (Estrada-Pérez et al., 2011). In order to precisely characterize the heat diffusion from the highly focused IR laser, we developed a two-photon fluorescent thermometry (TPFT) technique to measure the three-dimensional (3D) distribution of temperature rise in the region close to the laser focal point in water, 3% agarose (a tissue phantom of thermal conductivity similar to typical tissues) (Huang et al., 2004) and live zebrafish, respectively. The thermometry measures the temperature rise in tissues noninvasively based on the fluorescence signals of two fluorescent dyes (Appendix 3) (Estrada-Pérez et al., 2011; Natrajan and Christensen, 2008). In details, a temperature-sensitive dye (Figure 1), tetramethylrhodamine (TAMRA) which is conjugated with dextran, is adopted as the probe dye in TPFT, while fluorescein (FITC), which is insensitive to temperature and also conjugated with dextran (Figure 1C), is used as a reference dye to eliminate the fluctuation of probe dye fluorescence caused by a variety of interferences (Appendix 3). The temperature dependencies of fluorescence measured in pure TAMRA and FITC solutions are -0.882 ± 0.100%/°C and -0.165 ± 0.098%/°C respectively (Figure 1—figure supplement 2A and B, Figure 1—figure supplement 2—source data 1). The two-photon excited fluorescence (TPEF) intensity ratio of TAMRA and FITC is linearly correlated with the solution temperature due to large difference in temperature sensitivity between two dyes (Figure 1D and Figure 1—figure supplement 2). The temperature sensitivities of fluorescence intensity ratio are similar in water solution, 3% agarose and zebrafish in vivo (Figure 1D and Supplementary file 1a), indicating that the temperature coefficient of the fluorescent dextran remains stable in different environments. The high consistency between the actual temperature and the measured temperature (Figure 1E) demonstrates that TPFT can be used as an effective tool for in vivo measurement of the local temperature rise induced by IR laser heating in tissues. Single-cell labeling in zebrafish To study the dynamic change of temperature during IR laser heating, firstly we used TPFT to measure the temperature in water solution and 3% agarose with point heating. Low fluorescent dye concentrations were used to avoid self-absorption and fluorescence resonance energy transfer (FRET) (Figure 2—figure supplement 1 and Appendix 4). Using a high-sensitivity EMCCD as the spectra detector, the dynamic temperature change at the heating site can be recorded in real time. We found that the temperature at IR laser focal point increased sharply within 1 ~ 2 s after point heating and remained stable over the exposure time of the IR laser, before decreasing quickly to the ambient temperature as soon as the IR laser was turned off (Figure 2—figure supplement 2A). The depth of IR laser focal point in water and tissue phantom should be over 100 µm to minimize the thermal conduction at the intermedium surface (Figure 2—figure supplement 2B and C). Next, we measured the 3D temperature distributions (Figure 2A and B, Figure 2—source data 1), and found that high spatial resolution of heat shock could be achieved in 3% agarose. However, a large temperature gradient could not be built in water solution because of its high thermal conductivity and faster convection. Further, we conducted 3D temperature measurement in zebrafish in vivo and the results were compared with the measurement in agarose tissue phantom (Figure 2C and D, Figure 2—source data 1). Dextran-conjugated TAMRA and FITC were co-injected into fish embryos at one-cell stage. Then the embryos were raised to 1 day post fertilization (dpf). Muscle was chosen as the first tissue to perform temperature measurement because of its relatively uniform structure and simple microenvironment. To avoid laser-induced injury (Figure 2—figure supplement 3), we applied scan heating on zebrafish muscle and tissue phantom, in which the focused IR laser beam was scanned over an 8 µm × 8 µm region for 32 s instead of staying at a fixed heating point. Results showed that the thermal confinement in zebrafish muscle is higher than in tissue phantom, both laterally and axially (Figure 2C and D, Figure 2—source data 1). This indicates that the thermal conductivity of zebrafish muscle could be lower than that of 3% agarose. To visualize the thermal confinement clearly, we generated 3D view of the lateral temperature distributions based on the experimentally measured data (Figure 2E). As shown in Figure 2 E1 and E2, low thermal conductivity and inefficient convection of tissue phantom plays a critical role to confine the thermal energy and achieve single-cell resolution of heat shock. As shown in Figure 2 E2 and E3, there is no significant difference in the thermal confinement between point and scan heating methods. The results in Figure 2 E3 and E4 demonstrate that TPFT can finely evaluate the thermal distribution in zebrafish muscle in vivo and paves the way for evaluation of single-cell heat shock in different zebrafish tissues. Figure 2 with 4 supplements see all Download asset Open asset 3D temperature distribution measured through fluorescent thermometry and single-cell gene induction in zebrafish. (A) Lateral temperature distributions with IR laser heating in water and 3% agarose (95 mW IR laser was focused in samples for point heating). (B) Corresponding axial temperature distributions in water and 3% agarose. (C) Lateral temperature distributions with IR laser scan heating in 3% agarose and zebrafish muscle in vivo (95 mW IR laser was scanned in an 8 µm × 8 µm region during heating to avoid tissue injury). (D) Corresponding axial temperature distributions in 3% agarose and zebrafish muscle in vivo. Each statistical distribution curve in (A–D) is shown in terms of the mean with the standard deviation over more than five measurements. (E1–E4) 3D view of lateral temperature distributions with IR laser point and scan heating in water, 3% agarose and zebrafish muscle, respectively. (F1) A merged image of bright-field (gray) middle trunk, Dendra2-labeled nuclei (green) and a single myocyte expressing DsRedx via heat shock gene induction (red) in a one dpf zebrafish. (F2) A merged image of second harmonic generation (SHG) of muscle fibers (green) and the single myocyte expressing DsRedx (red). (G1) A merged image of bright-field (gray) hindbrain, GFP-labeled tyrosine hydroxylase-positive (th-positive) neurons (green) and a single neuron expressing DsRedx via heat shock gene induction (red) in a three dpf zebrafish. (G2) An enlarged image of (G1) with GFP and DsRedx-labeled neurons (by maximum projections). (H1) A merged image of bright-field (gray) middle trunk, DsRedx-labeled leukocytes (red) and a single cell expressing GFP via heat shock gene induction (green) in a two dpf zebrafish. (H2) An enlarged image of (H1) with merged DsRedx and GFP-labeled leukocytes. Arrows in (F–H): heat-shock labeled single myocyte, neuron and leukocyte, respectively. Scale bars: 50 µm (F1–H1); 30 µm (F2–H2). Figure 2—source data 1 3D temperature distribution in water, tissue phantom and zebrafish during IR laser heat shock. https://cdn.elifesciences.org/articles/52024/elife-52024-fig2-data1-v2.xlsx Download elife-52024-fig2-data1-v2.xlsx Next, we examined the single-cell labeling efficiency in vivo through the single-cell IR-LEGO system in various kinds of cells, including myocytes in the skeletal muscle, neurons in the brain, and coro1a+ leukocytes in the hematopoietic tissue at the aorta-gonad-mesonephros (AGM) and PBI of transgenic zebrafish (Figure 2F–H). Using TPFT, the temperature distribution in the heat shocked tissues was measured. With high IR laser power, the average temperature at the focal point (P00) of IR laser can reach as high as 50°C (Figure 2—figure supplement 4A, Figure 2—figure supplement 4—source data 1). Although considerable success rates of overall cell labeling can be achieved with this high temperature heat shock, the percentages of single-cell labeling within the overall labeling are relatively low (Supplementary file 1b). This is due to that effective heat shock gene induction can be induced with environmental temperature higher than 38°C (Shoji and Sato-Maeda, 2008) and the heating region of temperature over 38°C was greater than the single cell size. Therefore, in order to increase the efficiency of single-cell labeling, the IR laser power was then optimized to restrain the heat diffusion and limit the effective area of gene induction in a single-cell dimension. With the optimized heat shock condition, the temperature at 10 µm away from the focal point (P10) dropped below 38°C (Figure 2—figure supplement 4B and C, Figure 2—figure supplement 4—source data 1), preventing unwanted gene induction in neighboring cells. It was demonstrated that, after heat shock with optimized laser condition, successful overall cell labeling in myocytes, neurons and leukocytes can be observed in 36.7%, 18.6% and 50% of zebrafish, respectively, among which the efficiencies of single-cell labeling for these three types of tissues are 54.5%, 100% and 77.8%, respectively (Figure 2—figure supplement 4D and Supplementary file 1b). Compared with the high-temperature heat shock, the success rates of overall cell labeling with optimized heat shock condition were decreased, but the efficiency of single-cell labeling was significantly improved. This demonstrates that the single-cell IR-LEGO technology can efficiently and precisely induce heat shock-mediated gene editing within single cell in vivo in multiple tissues under optimized condition. Tracing single HE and progenies Similar to mammals, the definitive hematopoiesis of zebrafish initiates from the HE in the ventral wall of the dorsal aorta and was thought to give rise to hematopoietic stem cells (HSCs) (Bertrand et al., 2010; Kissa and Herbomel, 2010; Tian et al., 2017). Yet, our recent study has shown that in addition to generating HSCs, the HEs in the aorta also produces non-HSC progenitors capable of differentiating into T cells, myeloid and erythroid lineages but not B cells in a transient manner (Tian et al., 2017; Bertrand et al., 2007), highlighting the complexity of endothelial-hematopoietic transition (EHT), a process leading to the formation of blood stem and progenitor cells from the endothelium (Bertrand et al., 2010; Kissa and Herbomel, 2010). An important unsolved issue is whether these non-HSC-derived hematopoietic lineages, such as T lymphocytes, myeloid and erythroid cells, arise directly from distinct HE subpopulations that differentiate into different hematopoietic lineages independently or from a uniform HE population, which generates a common progenitor that further differentiates into multiple hematopoietic lineages. To address the issue, we applied the high-precision single-cell IR-LEGO technology to single HE lineage tracing in the PBI region where all three non-HSC-derived hematopoietic lineages but not HSCs are generated (Tian et al., 2017; Bertrand et al., 2007). Specifically, we estimated the physical sizes of HEs by measuring the distance between the nuclei of neighboring endothelial cells, and it shows that the average length of endothelial cells along the aortic floor in the PBI is 24.9 µm, with the minimum of 11.2 µm (Figure 3—figure supplement 1, Figure 3—figure supplement 1—source data 1). Given that we have successfully constrained the effective heat-shock region (>38°C) within 10 µm along the ventral wall of aorta (Figure 2—figure supplement 4B and Supplementary file 1b), it is highly feasible to label HE at single-cell resolution using our heat-shock microscope system. In order to label HEs and follow their fates, we generated a double transgenic Tg(kdrl:loxP-DsRedx-loxP-EGFP;coro1a:loxP-DsRedx-loxP-EGFP) fish, in which editable genetic reporter loxP-DsRedx-loxP-EGFP is under the control of endothelial-specific kdrl promoter (Jin et al., 2005) and leukocyte-specific coro1a promoter (Li et al., 2012), thus HEs and leukocytes (including myeloid and lymphoid cells) were marked by DsRedx (Figure 3A). The double reporter transgenic line was then outcrossed with Tg(hsp70l:mCherry-T2a-CreERT2) fish to obtain a triple transgenic Tg(kdrl:loxP-DsRedx-loxP-EGFP;coro1a:loxP-DsRedx-loxP-EGFP;hsp70l:mCherry-T2a-CreERT2) line (referred to as ‘triple Tg’ hereinafter) (Figure 3A). In this triple Tg embryo, single-cell IR-LEGO system-induced heat-shock and 4-OH tamoxifen (4-OHT) treatment would induce and subsequently activate CreER within one HE, resulting in the excision of the DsRedx cassettes. As a consequence, the targeted HE and its leukocyte progenies would be distinguished from the unlabeled cells by EGFP expression (Figure 3A). To ensure only one single HE was labeled in each embryo, we irradiated the 4-OHT treated embryos at 26–28 hpf (Figure 3B) prior to the initiation of EHT (Kissa and Herbomel, 2010; Tian et al., 2017), then immediately imaged the heat-shocked embryos and control embryos continuously to 48 hpf (Figure 3B and C, Videos 1 and 2). As shown in Figure 3C and Video 1, GFP signal began to emerge at ~6 hrs post heat-shock and 4-OHT treatment. The embryos with single GFP+ HE during 20 hrs of time-lapse imaging were then selected for hematopoietic lineage analysis and the contribution of the labeled HE to T lymphocytes and myeloid cells in each of these embryos were determined by counting the coro1a:GFP+ cells at 7 dpf (Supplementary file 1c and 1d). Because T lymphocytes are strictly located in the thymus at 7 dpf, the GFP signals in thymus were analyzed to determine the numbers of T lymphocyte derived from the labeled HE. Whole-mount fluorescent in situ hybridization and antibody staining verified that the GFP+ cells in the thymus were indeed rag1+ T lymphocytes (Figure 3—figure supplement 2A,B). Additionally, co-labeling of thymic epithelium and different hematopoietic cell types using specific transgenic zebrafish lines exhibited that T lymphocytes can also be effectively differentiated from other thymus-resident cells based on their small and round shapes (Figure 3—figure supplement 2C–F). Therefore, the numbers of GFP+ T lymphocyte in thymus can be accurately counted as the contribution of the labeled HE cell to lymphoid lineage (Figure 3D). On the other hand, the GFP+ cells distributed on the embryonic trunk are lyz+ or mpeg1+ myeloid cells (Figure 3D, Figure 3—figure supplement 3, Figure 3—figure supplement 3—source data 1). To avoid the disturbance by immature hematopoietic progenitors, GFP+ cells in the hematopoietic tissues including AGM, caudal hematopoietic tissue (CHT) and kidney were excluded from the myeloid cell statistics. Results showed that, albeit a small portion of zebrafish show GFP+ background signals in the control group (Figure 3E), the numbers of GFP+ T lymphocytes and myeloid cells in the single HE-labeled group were significantly higher compared with that in the control group (Figure 3E), demonstrating that the high-precision single-cell IR-LEGO system can efficiently trace the progenies derived from a single HE with high fidelity. Figure 3 with 3 supplements see all Download asset Open asset The high-precision IR-LEGO-mediated single HE lineage tracing. (A) The genetic labeling diagram of the lineage tracing experiment. The heat shock-induced CreERT2 line Tg(hsp70l:mCherry-T2a-CreERT2) is crossed with the double reporter fish Tg(kdrl:loxP-DsRedx-loxP-EGFP;coro1a:loxP-DsRedx-loxP-EGFP) to obtain triple transgenic Tg(kdrl:loxP-DsRedx-loxP-EGFP; coro1a:loxP-DsRedx-loxP-EGFP; hsp70l:mCherry-T2a-CreERT2) embryo. Upon IR laser illumination, the targeted HEs will express CreER. After 4-OHT treatment, the CreER will enter the cell nucleus and remove the DsRedx cassettes flanked by loxP from the genome. After that, the target HEs and their hematopoietic progenies will express GFP signal, thus be distinguished from the unlabeled DsRedx+ cells. (B) The work flow of the IR-LEGO-mediated single HE lineage tracing assay. (C) Representative time-lapse images of the single-HE heat-shocked embryos (upper row) and control embryos (lower row). Dotted lines depict the caudal aorta (CA) in the PBI region (dorsal wall on the top). Number at the top left corner in each image indicates the developmental stage of the embryos (hh:mm post fertilization). The heat-shock labeled HE (white arrow) on the ventral wall of CA gradually turns on the expression of GFP without affecting neighbor HEs, while the HEs in control embryos do not express GFP during time-lapse imaging. (D) Images of GFP+ T cells and myeloid cells in single-HE labeled fish and control fish at 7 dpf (back on the top). The left column shows small and round coro1a:GFP+ T cells in the thymus (depicted by dashed lines). The right column shows coro1a:GFP+ myeloid cells on the trunk, which have irregular shape. While GFP+ T cells and myeloid cells are persistently observed in many of the labeled fish (upper row), rare GFP sign

Extracellular vesicles (EVs) are released by most cell types but providing evidence for their physiological relevance remains challenging due to a lack of appropriate model organisms. Here, we developed an in vivo model to study EV function by expressing CD63-pHluorin in zebrafish embryos. A combination of imaging methods and proteomic analysis allowed us to study biogenesis, composition, transfer, uptake, and fate of individual endogenous EVs. We identified a subpopulation of EVs with exosome features, released in a syntenin-dependent manner from the yolk syncytial layer into the blood circulation. These exosomes are captured, endocytosed, and degraded by patrolling macrophages and endothelial cells in the caudal vein plexus (CVP) in a scavenger receptor- and dynamin-dependent manner. Interference with exosome biogenesis affected CVP growth, suggesting a role in trophic support. Altogether, our work represents a system for studying endogenous EV function in vivo with high spatiotemporal accuracy, demonstrating functional inter-organ communication by exosomes.

In this issue of Developmental Cell, Lin et al. (2019) identify in zebrafish skin macrophage-like cells that sample the environment through transepithelial protrusions and import antigen from the water for traditional tissue-resident macrophages. Remarkably, these "metaphocytes'' originate from the epidermis, challenging current assumptions about the lineage of tissue-resident macrophages.

Actin dynamics is central for cells, and especially for the fast-moving leukocytes. The severing of actin filaments is mainly achieved by cofilin, assisted by Aip1/Wdr1 and coronins. We found that in Wdr1-deficient zebrafish embryos, neutrophils display F-actin cytoplasmic aggregates and a complete spatial uncoupling of phospho-myosin from F-actin. They then undergo an unprecedented gradual disorganization of their nucleus followed by eruptive cell death. Their cofilin is mostly unphosphorylated and associated with F-actin, thus likely outcompeting myosin for F-actin binding. Myosin inhibition reproduces in WT embryos the nuclear instability and eruptive death of neutrophils seen in Wdr1-deficient embryos. Strikingly, depletion of the main coronin of leukocytes, coronin 1A, fully restores the cortical location of F-actin, nuclear integrity, viability, and mobility of Wdr1-deficient neutrophils in vivo. Our study points to an essential role of actomyosin contractility in maintaining the integrity of the nucleus of neutrophils and a new twist in the interplay of cofilin, Wdr1, and coronin in regulating F-actin dynamics.

Institut Pasteur, Unité Macrophages et Déve France. E-mail: emma.colucci@pasteur.fr CNRS, UMR 3738, Paris, France Nanotechnology Engineering Program, Ins Graduação e Pesquisa de Engenharia – CO Janeiro, Rio de Janeiro, 21941-972, Brazil Department of Anesthesiology, University o LCRB, CNRS, Université Paris 5, Sorbonne Departamento de Ciências Exatas e Educ Catarina, Blumenau, 89036-256, Brazil Institut Pasteur, Unité de Pathogénie M France Cite this: Chem. Sci., 2019, 10, 3654

Enhanced susceptibility to bacterial infection in the days following an acute virus infection such as flu is a major clinical problem. Mouse models have provided major advances in understanding viral-bacterial superinfections, yet interactions of the anti-viral and anti-bacterial responses remain elusive. Here, we have exploited the transparency of zebrafish to study how viral infections can pave the way for bacterial co-infections. We have set up a zebrafish model of sequential viral and bacterial infection, using sublethal doses of Sindbis virus and Shigella flexneri bacteria. This virus induces a strong type I interferons (IFN) response, while the bacterium induces a strong IL1β and TNFα-mediated inflammatory response. We found that virus-infected zebrafish larvae showed an increased susceptibility to bacterial infection. This resulted in the death with concomitant higher bacterial burden of the co-infected fish compared to the ones infected with bacteria only. By contrast, infecting with bacteria first and virus second did not lead to increased mortality or microbial burden. By high-resolution live imaging, we showed that neutrophil survival was impaired in Sindbis-then-Shigella co-infected fish. The two types of cytokine responses were strongly induced in co-infected fish. In addition to type I IFN, expression of the anti-inflammatory cytokine IL10 was induced by viral infection before bacterial superinfection. Collectively, these observations suggest the zebrafish larva as a useful animal model to address mechanisms underlying increased bacterial susceptibility upon viral infection.

Hematopoiesis leads to the formation of blood and immune cells. Hematopoietic stem cells emerge during development, from vascular components, via a process called the endothelial-to-hematopoietic transition (EHT). Here, we reveal essential biomechanical features of the EHT, using the zebrafish embryo imaged at unprecedented spatio-temporal resolution and an algorithm to unwrap the aorta into 2D-cartography. We show that the transition involves anisotropic contraction along the antero-posterior axis, with heterogenous organization of contractile circumferential actomyosin. The biomechanics of the contraction is oscillatory, with unusually long periods in comparison to other apical constriction mechanisms described so far in morphogenesis, and is supported by the anisotropic reinforcement of junctional contacts. Finally, we show that abrogation of blood flow impairs the actin cytoskeleton, the morphodynamics of EHT cells, and the orientation of the emergence. Overall, our results underline the peculiarities of the EHT biomechanics and the influence of the mechanical forces exerted by blood flow.

Macrophages infiltrate and establish in developing organs from an early stage, often before these have become vascularized. Similarly, leukocytes, in general, can quickly migrate through tissues to any site of wounding. This unique capacity is rooted in their characteristic amoeboid motility, the genetic basis of which is poorly understood. Trim33 (also known as Tif1-γ), a nuclear protein that associates with specific DNA-binding transcription factors to modulate gene expression, has been found to be mainly involved in hematopoiesis and gene regulation mediated by TGF-β. Here, we have discovered that in Trim33-deficient zebrafish embryos, primitive macrophages are unable to colonize the central nervous system to become microglia. Moreover, both macrophages and neutrophils of Trim33-deficient embryos display a reduced basal mobility within interstitial tissues, and a profound lack of a response to inflammatory recruitment signals, including local bacterial infections. Correlatively, Trim33-deficient mouse bone marrow-derived macrophages display a strongly reduced three-dimensional amoeboid mobility in fibrous collagen gels. The transcriptional regulator Trim33 is thus revealed as being essential for the navigation of macrophages and neutrophils towards developmental or inflammatory cues within vertebrate tissues.

Alphaviruses, such as chikungunya virus (CHIKV) and Sindbis virus (SINV), are vector-borne pathogens that cause acute illnesses in humans and are sometimes associated with neuropathies, especially in infants and elderly patients. Little is known about their mechanism of entry into the central nervous system (CNS), even for SINV, which has been used extensively as a model for viral encephalopathies. We previously established a CHIKV infection model in the optically transparent zebrafish larva; here we describe a new SINV infection model in this host. We imaged in vivo the onset and progression of the infection caused by intravenous SINV inoculation. Similar to that described for CHIKV, infection in the periphery was detected early and was transient, whereas CNS infection started at later time points and was persistent or progressive. We then tested the possible mechanisms of neuroinvasion by CHIKV and SINV. Neither virus relied on macrophage-mediated transport to access the CNS. CHIKV, but not SINV, always infects endothelial cells of the brain vasculature. By contrast, axonal transport was much more efficient with SINV than CHIKV, both from the periphery to the CNS and between neural tissues. Thus, the preferred mechanisms of neuroinvasion by these two related viruses are distinct, providing a powerful imaging-friendly system to compare mechanisms and prevention methods of encephalopathies.

By genetic linkage analysis in a large consanguineous Iranian family with eleven individuals affected by severe to profound congenital deafness, we were able to define a 2.8 Mb critical interval (at chromosome 1p21.2-1p21.1) for an autosomal-recessive nonsyndromic deafness locus (DFNB). Whole-exome sequencing allowed us to identify a CDC14A biallelic nonsense mutation, c.1126C>T (p.Arg376(∗)), which was present in the eight clinically affected individuals still alive. Subsequent screening of 115 unrelated individuals affected by severe or profound congenital deafness of unknown genetic cause led us to identify another CDC14A biallelic nonsense mutation, c.1015C>T (p.Arg339(∗)), in an individual originating from Mauritania. CDC14A encodes a protein tyrosine phosphatase. Immunofluorescence analysis of the protein distribution in the mouse inner ear showed a strong labeling of the hair cells' kinocilia. By using a morpholino strategy to knockdown cdc14a in zebrafish larvae, we found that the length of the kinocilia was reduced in inner-ear hair cells. Therefore, deafness caused by loss-of-function mutations in CDC14A probably arises from a morphogenetic defect of the auditory sensory cells' hair bundles, whose differentiation critically depends on the proper growth of their kinocilium.

The ontogeny of haematopoietic niches in vertebrates is essentially unknown. Here we show that the stromal cells of the caudal haematopoietic tissue (CHT), the first niche where definitive haematopoietic stem/progenitor cells (HSPCs) home in zebrafish development, derive from the caudal somites through an epithelial-mesenchymal transition (EMT). The resulting stromal cell progenitors accompany the formation of the caudal vein sinusoids, the other main component of the CHT niche, and mature into reticular cells lining and interconnecting sinusoids. We characterize a zebrafish mutant defective in definitive haematopoiesis due to a deficiency in the nascent polypeptide-associated complex alpha subunit (NACA). We demonstrate that the defect resides not in HSPCs but in the CHT niche. NACA-deficient stromal cell progenitors initially develop normally together with the sinusoids, and HSPCs home to the resulting niche, but stromal cell maturation is compromised, leading to a niche that is unable to support HSPC maintenance, expansion and differentiation.