Atsushi Kawakami

Previous studies have shown that tissue regeneration induces expression of genes that play important roles in regeneration. Recently, several studies have identified regeneration‐response enhancers (RREs) that activate gene expression by tissue injury. Particularly, we showed that RREs contain two transcription factor‐binding motifs: a bHLH transcription factor‐binding motif, an E‐box, and an AP‐1/bZIP transcription factor‐binding motif, a 12‐O‐Tetradecanoylphorbol 13‐acetate response element (TRE). However, the triggers and subsequent signals generated by injury are still unclear. In this study, we analyzed RRE activation using various injury models. Although inter‐ray incisions and skin exfoliation injuries did not activate RREs or regeneration genes, the fin puncture injury activated RREs and several regeneration‐response genes. After fin puncture injury, msxc was activated only on the proximal side of the hole where blastema‐like tissue was formed, whereas RREs, junbb, and fibronectin 1b (fn1b) were activated on both the proximal and distal sides, implying that activation of RREs, junbb, and fn1b is independent of blastema formation. Here, we also established a mild cryoinjury method. After this injury, transient vascular destruction, an increase in cell death, and an accumulation of myeloid cells were observed; however, no major morphological damage was observed. Importantly, msxc was not induced by cryoinjury, whereas fn1b, junbb, and 1.8 k RRE (−1.8 kb promoter of fn1b) were activated, suggesting that cryoinjury induces the responses of fn1b, junbb, and 1.8 k RRE without forming the blastema. Thus, our study shows that the cryoinjury model and the RRE transgenic (Tg) zebrafish may provide a useful platform for exploring injury signals.

Basal stem cells of the epidermis continuously differentiate into keratinocytes and replenish themselves via self-renewal to maintain skin homeostasis. Numerous studies have attempted to reveal how basal cells undergo differentiation or self-renewal; however, this has been hampered by a lack of robust basal cell markers and analytical platforms that allow single-cell tracking. Here, we report that zebrafish integrin beta 4 is a useful marker for basal cell labelling, irrespective of the body region, stage and regenerative status. We employed Cre-loxP recombination in combination with live cell tracking of single basal clones in the caudal fin and investigated the embryonic origin and behaviour of basal cells during fish growth and homeostasis. Although most basal cells, including those in fins, became quiescent in the adult stage, genetic cell ablation showed that basal cells were reactivated to either self-renew or differentiate, depending on the injured cell type. Our study provides a simple and easy-to-use platform for quantitative in vivo imaging of basal stem cells at wider stages and under various conditions.

Skeletal tissues including cartilage and bones are characteristic features of vertebrates that are crucial for supporting body morphology and locomotion. Studies mainly in mice have shown that osteoblasts and chondroblasts are supplied from several progenitors like the sclerotome cells in the embryonic stage, osteo-chondroprogenitors in growing long bones, and skeletal stem cells of bone marrow in the postnatal period. However, the exact origins of progenitor cells, their lineage relationships, and their potential to differentiate into osteoblasts and chondroblasts from embryos to adult tissues are not well understood. In this study, we conducted clonal cell tracking in zebrafish and showed that sox9a+ cells are already committed to either chondrogenic or osteogenic fates during embryonic stages and that respective progenies are independently maintained as mesenchymal progenitor pools. Once committed, they never change their lineage identities throughout animal life, even through regeneration. In addition, we further revealed that only osteogenic mesenchymal cells replenish the osteoblast progenitor cells (OPCs), a population of reserved tissue stem cells found to be involved in the de novo production of osteoblasts during regeneration and homeostasis in zebrafish. Thus, our clonal cell tracking study in zebrafish firstly revealed that the mesenchymal progenitor cells that are fated to develop into either chondroblasts or osteoblasts serve as respective tissue stem cells to maintain skeletal tissue homeostasis. Such mesenchymal progenitors dedicated to producing either chondroblasts or osteoblasts would be important targets for skeletal tissue regeneration. Skeletal tissues like cartilage and bones are characteristic features of vertebrates, but their cellular sources are not fully understood. We tracked the sox9a+ mesenchymal progenitor cells in zebrafish. The lineages that produce osteogenic or chondrogenic progenitors are fated during the early stages of animal life and kept independent throughout their life, even during regeneration.image

Injury triggers a genetic program that induces gene expression for regeneration. Recent studies have identified regeneration-response enhancers (RREs); however, it remains unclear whether a common mechanism operates in these RREs. We identified three RREs from the zebrafish fn1b promoter by searching for conserved sequences within the surrounding genomic regions of regeneration-induced genes and performed a transgenic assay for regeneration response. Two regions contained in the transposons displayed RRE activity when combined with the −0.7 kb fn1b promoter. Another non-transposon element functioned as a stand-alone enhancer in combination with a minimum promoter. By searching for transcription factor-binding motifs and validation by transgenic assays, we revealed that the cooperation of E-box and activator protein 1 motifs is necessary and sufficient for regenerative response. Such RREs respond to variety of tissue injuries, including those in the zebrafish heart and Xenopus limb buds. Our findings suggest that the fidelity of regeneration response is ensured by the two signals evoked by tissue injuries. It is speculated that a large pool of potential enhancers in the genome has helped shape the regenerative capacities during evolution.

The interaction between immune cells and injured tissues is crucial for regeneration. Previous studies have shown that macrophages attenuate inflammation caused by injuries to support the survival of primed regenerative cells. Macrophage loss in zebrafish mutants like cloche (clo) causes extensive apoptosis in the regenerative cells of the amputated larval fin fold. However, the mechanism of interaction between macrophage and injured tissue is poorly understood. Here, we show that a phosphoinositide 3-kinase gamma (PI3K gamma)-mediated signal is essential for recruiting macrophages to the injured tissue. PI3K gamma inhibition by the PI3K gamma-specific inhibitor, 5-quinoxalin-6-ylmethylene-thiazolidine-2,4-dione (AS605240 or AS), displayed a similar apoptosis phenotype with that observed in clo mutants. We further show that PI3K gamma function during the early regenerative stage is necessary for macrophage recruitment to the injured site. Additionally, protein kinase B (Akt) overexpression in the AS-treated larvae suggested that Akt is not the direct downstream mediator of PI3K gamma for macrophage recruitment, while it independently plays a role for the survival of regenerative cells. Together, our study reveals that PI3K gamma plays a role for recruiting macrophages in response to regeneration.

Injury triggers genetic program to induce gene expression for regeneration. Several studies have recently reported the identification of regeneration-response enhancers (RREs) in zebrafish; however, it remains unclear whether a common mechanism operates in RREs. Here, we show that E-box and activator protein 1 (AP-1) motifs cooperatively function as RREs. We identified three RREs from the fn1b promoter by a search of conserved sequences and an in vivo transgenic assay for regeneration-response in zebrafish. Two of them derived from transposons displayed RRE activity only when combined with the −0.7 kb fn1b promoter, while another non-transposable element functioned as a standalone enhancer. A search for transcription factor-binding motifs and validation by transgenic assay revealed that both of E-box and AP-1 motifs are necessary and sufficient for RREs. Such RREs responded to variety of tissue injuries including zebrafish heart and Xenopus limb bud regenerations. Our findings highlight that regeneration is regulated by merging two activating signals evoked by tissue injuries. It is speculated that a large pool of potential enhancers in the genome shaped regenerative capacities during evolution. SUMMARY STATEMENT: The study revealed that regeneration-response enhancer is composed of two transcription factor-binding motifs. The fidelity of regeneration-dependent gene expression is ensured by merging two activating signals evoked by injuries.

Versican is an evolutionary conserved extracellular matrix proteoglycan, and versican expression loss in mice results in embryonic lethality owing to cardiovascular defects. However, the in utero development of mammals limits our understanding of the precise role of versican during cardiovascular development. Therefore, the use of evolutionarily distant species that develop ex utero is more suitable for studying the mechanistic basis of versican activity. We performed ENU mutagenesis screening to identify medaka mutants with defects in embryonic cardiovascular development. In this study, we described a recessive point mutation in the versican 3′UTR resulting in reduced versican protein expression. The fully penetrant homozygous mutant showed termination of cardiac development at the linear heart tube stage and exhibited absence of cardiac looping, a constricted outflow tract, and no cardiac jelly. Additionally, progenitor cells did not migrate from the secondary source towards the arterial pole of the linear heart tube, resulting in a constricted outflow tract. Furthermore, mutants lacked blood flow and vascular lumen despite continuous peristaltic heartbeats. These results enhance our understanding of the mechanistic basis of versican in cardiac development, and this mutant represents a novel genetic model to investigate the mechanisms of vascular tubulogenesis.

Hypertrophic cardiomyopathy (HCM) is a hereditary disease characterized by cardiac hypertrophy with diastolic dysfunction. Gene mutations causing HCM have been found in about half of the patients, while the genetic etiology and pathogenesis remain unknown for many cases of HCM. To identify novel mechanisms underlying HCM pathogenesis, we generated a cardiovascular-mutant medaka fish non-spring heart ( nsh ), which showed diastolic dysfunction and hypertrophic myocardium. The nsh homozygotes had fewer myofibrils, disrupted sarcomeres and expressed pathologically stiffer titin isoforms. In addition, the nsh heterozygotes showed M-line disassembly that is similar to the pathological changes found in HCM. Positional cloning revealed a missense mutation in an immunoglobulin (Ig) domain located in the M-line-A-band transition zone of titin. Screening of mutations in 96 unrelated patients with familial HCM, who had no previously implicated mutations in known sarcomeric gene candidates, identified two mutations in Ig domains close to the M-line region of titin. In vitro studies revealed that the mutations found in both medaka fish and in familial HCM increased binding of titin to muscle-specific ring finger protein 1 (MURF1) and enhanced titin degradation by ubiquitination. These findings implicate an impaired interaction between titin and MURF1 as a novel mechanism underlying the pathogenesis of HCM.

It has been proposed that cells are regulated to form specific morphologies and sizes according to positional information. However, the entity and nature of positional information have not been fully understood yet. The zebrafish caudal fin has a characteristic V‐shape; dorsal and ventral fin rays are longer than the central ones. This fin shape regenerates irrespective of the sites or shape of fin amputation. It is thought that reformation of tissue occurs according to positional information. In this study, we developed a novel transplantation procedure for grafting a whole fin ray to an ectopic position and examined whether the information that specifies fin length exists within each fin ray. Intriguingly, when long and short fin rays were swapped, they regenerated to form longer or shorter fin rays than the adjacent host fin rays, respectively. Further, the abnormal fin ray lengths were maintained for a long time, more than 5 months, and after further re‐amputation. In contrast to intra‐fin grafting, when fin ray grafting was performed between fish, cells in the grafts disappeared due to immune rejection, and the grafted fin rays adapted to the host position to form a normal fin. Together, our data suggest that the information that directs fin length does exist in cells within a single fin ray and that it has a robust property—it is stable for a long time and is hard to rewrite. Our study highlighted a novel positional information mechanism for directing regenerating fin length.