The Hippo signalling pathway has been evolutionary conserved and shown to play a key role in controlling organ size through regulating the balance between cell proliferation and cell death. Also, in mammals the Hippo pathway acts as a tumor suppressor. When the Hippo pathway gets activated the transcriptional coactivator Yorkie (known as Yap in mammals) is repressed. On the other side, when the Hippo pathway is inactive, Yorkie is active. Yorkie induces cell proliferation and inhibits apoptosis and, therefore, its silencing results in a decrease in cell proliferation and increase in cell death.
A recent paper from the laboratory of Eugene Berezikov reports on the function of the Hippo pathway in the flatworm Macrostomum lignano (http://www.ncbi.nlm.nih.gov/pubmed/23495768). Similarly to freshwater planarians, M. lignano is a growing model to study stem cell-based regeneration. All these different flatworms are an excellent model in which to study the role of signalling pathways on stem cell regulation in vivo. In this study, the authors first isolated the homologues in M. lignano of the core components of the Hippo pathway including Hippo (Hpo), Salvador (Sav), Warts (Wts), Mats and Yorkie (Yki/Yap). Whole mount in situ hybridizations showed that in adult animals Hpo, Sav, Wts and Mats show similar expression patterns, being enriched in gonads and also found in neoblasts (pluripotent stem cells) and differentiated tissues. On the other hand, Yap is expressed specifically in the gonads and neoblasts as its expression is completely abrogated after irradiation (a way to eliminate all the neoblasts and dividing cells in flatworms).
Then the authors moved forward to functionally characterize all these genes first in adult homeostatic animals and then during regeneration. In adults, 2 weeks after RNAi treatment to silence Hpo, Sav, Wts and Mats the animals start to develop small outgrowths along the body. After 4-6 weeks there is an increase in the size of these outgrowths and epidermal bulges form throughout the body. Finally, all animals die within 8 weeks of treatment. In contrast, the silencing of Yap results in head regression followed by ventral curling and lysis of the whole animal, a phenotype that resembles those obtained in freshwater planarians when they are depleted of neoblasts. Compared to freshwater planarians M. lignano shows more limited regenerative abilities; however, when amputated behind the head they can regenerate a new posterior part. After Hpo(RNAi) animals regenerate a new posterior part but, remarkably, this new part appears to be larger than in controls. Moreover, the animals also develop bulges around the cutting site and show other morphological aberrations and disrupted allometric scaling. However, cell differentiation does not seem to be impaired in them. All these morphological phenotypes correlate with an increased cell proliferation detected in the blastemas of these animals after silencing Hpo. In contrast the silencing of Yap results in a significant decrease in the number of proliferating cells within the blastema, which blocks the regeneration of a new posterior region. In few days all the animals die.
Finally, the authors used BrdU to label neoblasts after silencing Hpo and Yap in adult homeostatic animals. By day 10 of treatment the number of S-phase cells is significantly higher in Hpo(RNAi) and significantly lower in Yap(RNAi), compared to controls. By day 20 not only the number of S-phase cells was much higher after Hpo(RNAi) but also were found all throughout the body, compared to the more restricted spatial distribution found in controls.
In summary, the results reported in this paper clearly show that the Hippo pathway is conserved in the flatworm M. lignano and, more importantly, that plays a key role in regulating neoblast biology during homeostasis and regeneration. Similarly to mammals the Hippo pathway acts as a tumor suppressor in these flatworms as Hpo(RNAi) animals develop outgrowths and bulges all throughout the body. Also, Yap appears to be required to maintain neoblasts self-renewal, which agrees with the role of Yap in maintaining the pluripotency of mammalian embryonic and induced stem cells. Further studies in M. lignano could help to characterize upstream and downstream elements of this pathway to better understand how the Hippo pathway regulates the size of the regenerating parts by controlling stem cell proliferation, cell death and differentiation.
Among vertebrates, zebrafish are remarkable animals as they can regenerate their brains. A recurrent topic in this blog is how the inflammatory response modulates the regenerative versus the non-regenerative output after an injury in different contexts. Whereas inflammation appears to have an inhibitory effect on mammalian neural regeneration (by promoting glial scar formation) and limiting cell proliferation, survival and migration), it does not seem to play such negative effect during zebrafish brain regeneration. On the contrary, a paper from the laboratory of Michel Brand reports not only that inflammation is not inhibitory but is required to activate brain regeneration in zebrafish (http://www.ncbi.nlm.nih.gov/pubmed/23138980).
First, the authors observed how after a brain injury there is a fast inflammatory response with an increase in the number of microglia and leukocytes and the expression of the proinflammatory cytokines IL-8, IL-1b and TNF-a. Then, the authors sought to determine whether this inflammatory response activates radial glial cells. What they did was to experimentally dissect the traumatic brain injury from the inflammatory response by injecting zymosan A in order to induce inflammation without injuring. After zymosan A delivery the induced inflammation detected pretty much mimics the inflammatory response after brain injury. Thus, for example, this induced inflammation results also in an increase in radial glial cells proliferation and their posterior neural differentiation. Remarkably, using an anti-inflammatory drug to immunosuppress the zebrafish significantly reduced the proliferation and neurogenetic response in the lesioned brains.
So far, these results indicate that inflammation is implicated in activating proliferation and neurogenesis. Next, the authors carried a transcriptome screen to identify the players in this process. They identified a cysteinyl leukotriene receptor (cysltr1) being induced after brain injury, predominantly in radial glial cells. Leukotrienes are a family of inflammatory mediators produced from arachidonic acid. In the zebrafish brain, the expression of cysltr1 is upregulated after zymosan A injections. Remarkably, the use of an antagonist of cysltr1 reduces radial glial cells proliferation and neurogenesis after injury. In order to better characterize the role of leukotriene signaling in this process they tested the effects of injecting leukotriene C4 (LTC4), a ligand of cysltr1. The injection of LTC4 into a non-injured brain results in a significant increase in reactive proliferation and neurogenesis. Finally, as the adult zebrafish brains are continuously making new neurons the authors wanted to determine if inflammation enhances neurogenesis by amplifying the existing signals in those adult brains or, alternatively, initiates a specific injury-induced program. To answer this question they analyzed the expression of gata3, a gene that is specifically induced after a brain injury but is not detected under homeostatic conditions. Zymosan A injections are sufficient to induce gata3 upregulation and this expression is silenced when zymosan A is injected in immunosuppressed animals. On the other side, LTC4 injections activate also gata3 expression that, again, is silenced when cysltr1 is blocked. Overall, the authors conclude that the inflammatory response, including LTC4 signalling, initiates a specific neurogenesis regeneration program.
In summary, and although there are contradictory studies about the positive and negative effects of acute inflammation on wound healing this study clearly couples the inflammatory response to cell proliferation and neurogenesis in a regeneration-competent model. This may help to better understand the molecular players required for a successful regeneration in zebrafish and, on the other side, it also opens the door to future therapeutic applications in regeneration-non-competent animals.
Several times in this blog I have pointed out the importance to extend our regenerative studies to more different species in order to have a broader view of this fascinating process as well as for a more comparative approach. Platyhelminthes are amazing animals not only because many of them show very high regenerative capabilities but also because these are based upon the presence of adult pluripotent stem cells. Among Platyhelminthes, model species include the freshwater planarians Schmidtea mediterranea and Dugesia japonica as well as the marine flatworm Macrostomum lignano.
A report from Ulrich Dirks and Jörg A. Ott has started to characterize the proliferative response associated to the regenerative process in the Catenulida, Paratenula galateia (http://www.ncbi.nlm.nih.gov/pubmed/22729484). Catenulida represent the most basally branching group of the Platyhelminthes. P. galateia is an interesting free-living animal with no mouth and gut, that reproduces asexually by paratomy and that harbors intracellular bacterial symbionts. These animals can be divided into an anterior rostrum and the trophosome region where the symbionts are found. In the rostrum these animals have a brain from which rostral nerves extend up to the most anterior tip. Two main longitudinal nerves extend from the rostrum towards the posterior regions of the animal. S-phase cells localize mostly throughout the trophosome up to behind the brain, so not dividing cells are seen in the most anterior part of the rostrum. When amputated behind the brain, the trophosome is able to regenerate a new rostrum; however, the rostrum is not able to regenerate the trophosome.
The authors analyse the dynamics of neoblast proliferation by combining EdU and BrdU pulses during rostrum regeneration. By 48h of regeneration the wound is closed by the constriction of circular muscles and the flattening of epidermal cells. At this stage EdU and BrdU positive cells are evenly distributed in the whole trophosome fragment. On the other side, stainings with an anti-serotonin antibody show the truncated longitudinal cords in the wound area. After 5 days of regeneration there is strong accumulation of proliferative cells within the forming blastema. Interestingly, an accumulation of proliferating cells is also found along the longitudinal nerve cords. At this stage, however, the amputated nerve cords appear still truncated and not extending into the blastema. The rest of the body shows an even distribution of low density of proliferating cells. Then, around 7 days of regeneration the new tip of the rostrum starts growing and a strong accumulation of proliferating cells is still observed in the blastema. Also, a prominent commissure appears at the anterior end of the nerve cords and some neuronal processes extend anteriorly, possibly corresponding to the regenerating rostral nerves. Finally, regeneration appears to be practically completed by day 11 after amputation. On the other side, all the rostrum fragments die after a couple of weeks without any sign of posterior regeneration.
Thus, this study represents a first step in trying to determine the proliferative dynamics of the neoblasts during regeneration in this Catenulida. However, and as the authors state, a more detailed time course of the regenerating process in terms of the proliferative response of the neoblasts is needed. Also, it will be interesting to further investigate how these neoblasts within the blastema exit the cell cycle and differentiate into the different cell types. Finally, it would be also interesting to study whether the failure to regenerate posterior regions from a rostrum fragment is caused by the low number of neoblasts present in the rostrum or the fact that the rostrum lacks enough symbionts for its nutrition. Maybe it would be interesting to analyze, in case the authors have not done it yet, what happens when the amputation is done not behind the brain but through the middle region of the trophosome. Would then the anterior fragment bearing the rostrum and part of the trophosome be able to regenerate the posterior end?
In previous posts I have commented on the regulatory role that the inflammatory and immune responses may have during regeneration. Whereas in many non-regenerative models wound healing is followed by the formation of a fibrotic scar that blocks further regeneration, scar-free wound healing is a rather conserved feature in regeneration-competent species. It is known for example that, in some contexts, the mammalian embryo is able to heal wounds without forming a fibrotic scar and, consequently, allowing a functional regeneration. During mammalian foetal development this capacity of scarless wound healing is lost as the immune system develops. Some studies have reported that the inhibition of anti-inflammatory factors in mice embryos impairs their capacity to heal wounds without making a fibrotic scar.
A paper from the laboratory of Nadia Rosenthal provides evidences of the requirement of macrophages for successful limb regeneration in salamanders (http://www.ncbi.nlm.nih.gov/pubmed/23690624). Macrophages play important roles as phagocytes in order to remove cell debris after injury as well as acting as a source of pro- and anti-inflammatory factors and growth factors that promote cell migration and proliferation. By checking the expression of a collection of cytokines the authors observed the rapid induction of cytokines, chemokines and inflammatory markers after 1 day of limb regeneration. But they also observed high levels of anti-inflammatory and anti-fibrogenic cytokines at this very early stage. In mammalian wound healing these anti-inflammatory cytokines are induced at later stages of wound healing. The authors then focussed in this early induction of anti-inflammatory factors. First they describe how after just 1 day of regeneration monocytes and macrophages accumulate around the wound region, with their numbers peaking by day 4-6 of regeneration. Next, they used a method that allows the transient in vivo ablation of the macrophages and analysed limb regeneration in that macrophage-free context. Remarkably, when macrophages are eliminated during the early stages of regeneration (before the blastema is formed), limb regeneration is completely blocked. During the first week of regeneration, and before the macrophages return, they observed an increased level of inflammatory cytokines and a reduced level of anti-inflammatory cytokines. In addition, they detected a decrease in the expression of MMP9 and MMP3, two matrix metalloproteinases with an important function for remodelling of the extracellular matrix, several blastema and dedifferentiation markers and the TGF-beta pathway. After macrophage ablation proliferation is also strongly reduced within the stump mesenchyme. Therefore, the authors conclude that depleting the macrophages before amputation disrupts specific gene pathways required for the progression from wound healing to regeneration. In addition, after macrophage depletion those animals displayed fibrotic stumps with extensive collagen I and IV deposition compared to controls, which suggests that macrophages are also regulators of these matrix degrading enzymes. This scar tissue became permanent even after 90 days, despite macrophage repopulation takes place after 2 weeks of treatment. Remarkably, the re-amputation of such limb stumps after macrophage repopulation results in a successful regeneration of a limb.
In summary the authors propose that in salamanders the macrophages are required to orchestrate the early response to injury and activate the limb regeneration program, probably by promoting cell dedifferentiation and the proliferation of progenitor cells needed to rebuild the missing tissues. It remains for future studies whether the early induction of anti-inflammatory factors could be used as therapy for promoting scar-free healing in non-regenerating animals and the subsequent activation of a successful regenerative program.