Newts, urodelian amphibians as axolotls, show a great regenerative capacity compared to most vertebrates. Similarly to axolotls, that are really paedomorphic retaining larval features, truly adult newts can regenerate limbs, lenses, the heart and CNS neurons, for example. Due to its enormous genome size (about ten times the size of human genome), however, previous attempts to have a sequenced genome or reference transcriptome had not been very successful. But things look very different now as the laboratories of Thomas Braun and his collaborators have just published a de novo assembly of the transcriptome of Notophtalmus viridescens, a newt model species for regeneration (http://www.ncbi.nlm.nih.gov/pubmed/23425577). The authors combined several sequencing approaches and platforms to end up with around 38,000 annotated putative transcripts, from which about 15,000 have been validated, by proteomics, to code for proteins. Total RNA was obtained from a variety of uninjured tissues as well as from regenerating hearts, limbs and tails at different timepoints. Also, RNA was obtained from different developmental stages with the purpose of getting as many transcripts as possible.
One of the most surprising results after the bioinformatic analyses of such amount of data has been the identification of new protein families, some of them urodele-specific and some of them newt-specific. Thus, the authors report 583 protein-coding transcripts with no hit in public databases and 243 protein-coding transcripts similar to urodele proteins only. Expression analyses (by microarrays and RT-PCR) of some of these genes during the regeneration of different organs, such as the heart and lens, show that they display specific expression changes during the regenerative process (either upregulation or downregulation). These results, as the authors discuss, may provide some light over the debate on whether the regenerative capacity has appeared convergently many times in distinct lineages or is in fact a basal feature of metazoans that has been, otherwise, lost in many other lineages. Although it seems clear that newts may have novel protein families the question that needs to be solved next is whether or not these proteic novelties are fundamental for the amazing regenerative capacities of newts compared to other vertebrates, including other amphibians. So, and it has been pointed out in previous comments to some posts of this blog, once the “omics” approach is done it is time then to go back to the real animal and check in vivo the function of all these putative interesting genes. Also, here, it will be fundamental to compare the function of orthologues in different species with similar or different regenerative capacity. For example, the authors show here how some newly identified genes show specific expression changes in dorsal and ventral iris during lens regeneration (by RT-PCR). It could be then interesting to go deeper and characterize the expression of these genes by in situ hybridization during lens regeneration and compare it to, for example, what happens during axolotl lens regeneration. As said in a previous post, whereas newts regenerate their lenses from the dorsal iris, axolotls can regenerate them from both dorsal and ventral iris (only at a specific developmental stage).
Although the idea that species that regenerate can do it because they have a “magical regeneration gene” (not present in non-regenerating animals) can be somehow appealing, the reality is that in most classical models of regeneration there are not really many proved examples yet of these magic species-specific factors. One of these genes, found in newts, is Prod1 that belongs to a family of three-finger protein (TFP) specific to salamanders and that is required for limb regeneration.
But looking at the phylogenetically wide distribution of regeneration as a trait, it seems reasonable also to consider that regeneration is probably an ancestral trait already present at the base of metazoans. That would account for the fact that animals that regenerate share key features for a successful regeneration such as for example a scarless wound healing and the capacity to re-activate conserved molecules and pathways that are only active during embryonic development in those species with poor regenerative capabilities. On the other hand, however, it could be also true, as it may happen in newts, that upon these conserved basic properties different lineages may have incorporated specific elements such as for example Prod1. Thus, in newts Prod1 exerts its function by interacting with other well conserved elements such as nAG or EGFR (epidermal growth factor receptor). In fact Prod1, as well as other TFP members are quite interesting from an evolutionary point if view because not only is Prod1 different in newts and axolotls (GPI-anchored in newts and secreted in axolotls), but also there are other examples of TFPs that appear specific of certain taxons. Another taxon-specific TFPs include some required for the formation of venom apparatus in elapid snakes and a GPI-anchored protein in Drosophila that binds to the shaker potassium channel.
So, maybe in evolutionary terms regeneration is an ancestral trait that in different lineages may have incorporated into its molecular programme taxon-specific elements to promote, trigger or enhance this amazing process. Newts, as suggested by the data presented in this paper, can become an excellent model in which to analyze whether its amazing regenerative abilities depend upon these novel identified proteins and how these interact with well-conserved molecules and pathways that are necessary to be reactivated during regeneration.
Planarian regeneration depends on a unique population of pluripotent stem cells called neoblasts. When a planarian is amputated neoblasts close to the wound respond by actively proliferating and migrating towards the wound forming the regenerative blastema in which most of the missing structures will differentiate. Although the pluripotency of the neoblasts as a whole population has been classically suggested and believed, it was in 2011 that the laboratory of Peter Reddien provided strong experimental evidence that at least a part of the neoblast population includes truly pluripotent stem cells. Thus, a single of these pluripotent neoblasts can rescue planarians lethally-irradiated (http://www.ncbi.nlm.nih.gov/pubmed/21566185).
The regenerative blastema has been classically seen as a rather uniform mass of undifferentiated cells, probably multipotent. However, recent data suggest that this could be not entirely true. In a recent hypothesis paper, Reddien discusses this possibility as well as at what stage of regeneration neoblasts and/or neoblast progeny are specified to differentiate into the required cell types (http://www.ncbi.nlm.nih.gov/pubmed/23404104). The author proposes two alternative (although not necessarily mutually exclusive) models: in the “naïve neoblast model” pluripotent neoblasts divide and migrate into the blastema where they would stop dividing and then be specified and differentiated into the proper cell types; alternatively, in the “specialized neoblast model”, there would be specialized neoblasts that would produce different lineage-committed non-dividing blastema cells. In this last case, therefore, the specification of the blastema cells towards a particular cell fate would occur somehow prior they enter the blastema. The “specialized neoblast model” is supported by the fact that at least for three different cell types: eye-pigment, photoreceptors and protonephridia, specific cell populations of progenitor-like neoblasts express specific transcription factors that specify those neoblasts towards those cell fates (http://www.ncbi.nlm.nih.gov/pubmed/21852957, http://www.ncbi.nlm.nih.gov/pubmed/21937596). Still, they are “neoblasts” according to the definition of neoblast used here: “cells expressing smedwi1 and irradiation sensitive (during the first 24h after irradiation)”.
There is no doubt that those recent findings of progenitor cells for some planarian cell types have re-opened an exciting debate on how heterogeneous are neoblasts in terms of their potentiality. As stated in Reddien’s paper many questions need to be solved, starting with a definition of a neoblast. Also, it would be important to determine up to what extend these specialized neoblasts are not only present in uninjured animals but divide and are maintained in them for regular cell turnover. If specialized neoblasts exist at low numbers in uninjured animals what happens after amputation: do naïve multipotent neoblasts make more specialized neoblasts? Do the few specialized neoblasts divide and make more of them? Or both?
Finally, as both models make an important distinction on whether blastema cells are specified within (naïve model) or outside (specialized model) the blastema, it would be then appropriate to look for a good definition of a blastema. Very often a planarian blastema is defined as the mass of unpigmented tissue that grows below the wound, but body pigmented cells differentiate relatively late so after 5 days of regeneration you have a rather unpigmented mass of cells but in fact inside a new brain, for example, has already differentiated. So, up to what stage during regeneration a blastema is a blastema? And also, how early can we talk about a blastema? Many of the important decisions such as, for example, the establishment of proper polarity are made very early after amputation, and before any sign of new tissue really appearing below the wound. And that polarity decisions are going to affect the cell types you will need; it is not the same regenerating a head with its brain than a tail. So, during regeneration, the signals that generate specialized neoblasts from naïve pluripotent neoblasts where do they come from? From pre-existing cells in the stump? When the blastema is formed, does it have any role in cell specification (in the “specialized neoblast model”)? or does it play a mere organizational role? Anyway the definition of “blastema” would probably deserve another post by itself.
During regeneration not only missing tissues and structures are re-made but also they need to be precisely patterned and integrated along the preexisting defined body axes. Thus, in the case of axolotl limb regeneration, if you amputate at the level of the wrist, only the hand is regenerated. If you amputate through the lower arm, then a lower arm and a hand are regenerated. Finally, if you amputate at the level of the upper arm, a new upper arm, lower arm and hand regenerate. That is, only the structures distal to the amputation plane are regenerated. This is known as the rule of distal transformation. More amazingly, classical experiments showed that if a hand blastema is grafted onto a stump at the level of the upper arm, a normal limb regenerates with the hand blastema cells contributing only to the new hand. In 2009, the laboratory of Elly Tanaka published a very important paper showing that contrary to the notion of a blastema being composed by a mass of undifferentiated cells, a more realistic view is that of a heterogeneous blastema formed by different populations of lineage-restricted progenitor cells (http://www.ncbi.nlm.nih.gov/pubmed/19571878). In that paper they determined that during regeneration new muscle, Schwann and epidermal cells derive from preexisting muscle, Schwann and epidermal cells, respectively. New dermis and skeletal cells could derive from either pre-existing dermis or skeletal cells. The main message was then that when cells dedifferentiate and proliferate in order to provide the blastema with new cells to regenerate the missing structures they somehow keep a memory of their origin. Then, combining specific GFP labeling and grafts for different cell types and tissues they showed that cartilage cells obey the rule of distal transformation but that Schwann cells do not.
Now, a recent paper by Nacu and collaborators from Tanaka’s lab have expanded the characterization of which cell types and tissues obey this rule (http://www.ncbi.nlm.nih.gov/pubmed/23293283). Thus, when grafting a wrist blastema from an animal in which connective cells express GFP onto an upper arm stump of a non-labeled host, GFP positive connective cells are found only in the regenerated hand. But when the authors transplant an upper arm blastema from an animal in which connective cells express GFP onto an upper arm stump of a non-labeled host, now GFP positive cells are found in the regenerated upper arm, lower arm and hand. This indicates that connective tissue cells obey the rule of distal transformation. However, when using GFP labeled muscle cells they saw that these do not obey that rule. So, if a wrist blastema from an animal in which muscle fibers express GFP is transplanted onto an upper arm stump of a non-labeled host, GFP muscle cells are found in the upper arm, lower arm and hand. In addition to these results, the authors also provide evidences that similarly to what happens during embryonic development the β-catenin pathway appears to have a role during the patterning of the regenerated muscle tissues. This provides a new example of how some processes are regulated by the same molecules and pathways in both development and regeneration.
In summary these results may help to characterize the main actors and processes involved in the establishment of the proximo-distal outcome of axolotl limb regeneration.
Adult neurogenesis has been found in many vertebrates although different groups of animals exhibit different degrees of limitations in terms of where in the CNS adult neurogenesis takes place, how many new neurons in normal unperturbed conditions are made and whether or not this neurogenesis can be triggered or enhanced somehow after any injury or in neurodegenerative diseases. Thus, for example, fish, amphibians and reptiles show a persistent neurogenesis throughout adulthood and can respond to brain injuries by inducing a variable regenerative response. In mammals and birds, adult neurogenesis appears mainly restricted to the olfactory bulb and hippocampus and the vocal center, respectively; however, a regenerative or repair response in the mammalian CNS is quite limited.
Among vertebrates, amphibians are the real champions of regeneration and that includes the capacity to regenerate neuronal populations in the CNS. Thus, works from the laboratory of András Simon in newts have shown that a regenerative response can occur from the activation of cells located in normally quiescent brain regions (http://www.ncbi.nlm.nih.gov/pubmed/21068061), and this way dopaminergic neurons from the midbrain can be completely regenerated after ablation, being this process regulated by dopamine itself (http://www.ncbi.nlm.nih.gov/pubmed/21474106). Now, a paper by Malcolm Maden and collaborators report on the regeneration of the axolotl brain (http://www.ncbi.nlm.nih.gov/pubmed/23327114). The authors describe the presence of proliferating endogenous neural progenitors throughout the ventricular zone of the adult brain, especially in the telencephalon and cerebellum and more reduced in other brain regions. Upon removal of telencephalon segments there is an upregulation of the proliferation of progenitor cells from the ventricular zone of the damaged region. The telencephalon is then effectively regenerated after several weeks. Interestingly, the results presented here appear to confirm the idea that the olfactory nerve may provide an important cue to stimulate brain regeneration. Thus, if the anterior third of the telencephalon is removed severing the olfactory nerve, the telencephalon does not regenerate until the olfactory nerve re-grows and contacts the remaining telencephalon. What is the nature of such stimulating cue is something that still needs to be addressed.
The extracellular matrix (ECM) does not only provide with a physical or structural support to cells and tissues but also it is a source of signaling and regulatory functions that impact most biological functions. Consequently, regeneration is also a process in which the ECM may obviously play an important role. Thus, for instance, just think on freshwater planarians. During regeneration and homeostasis stem cells proliferate and migrate towards the tissues or structures in which new cells are needed (i.e. the blastema or an old organ). Also, during these processes, the planarians may go through an extensive remodeling of the preexisting tissues. Considering the relevance that the ECM appears to have in regulating cell adhesion, cell migration and stem cell niche, self-renewal or differentiation in other systems, it is surprising how little we know about the role of ECM during planarian regeneration, or even how it changes during the required remodeling associated to regeneration or homeostasis. Now, a recent paper by Isolani and collaborators from the laboratory of Renata Batistoni reports the functional characterization of four matrix metalloproteinases (MMPs) in these animals (http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0055649).
Surprisingly, the silencing of two of these MMPs (mmp1, a zinc-dependent proteinase and mt-mmpA, a membrane-anchored proteinase) results in very severe (leading to lethality) phenotypes affecting normal homeostasis in intact planarians, but without apparently impairing regeneration in a significant way. This is true especially for mmp-1, whereas the knockdown of mt-mmpA delays the normal growth and differentiation of the blastema. The silencing of mmp-1 in non-regenerating animals results in tissue disorganization, breaking of the basal lamina and multilayered epidermis. At the cellular level, and compared to controls, the number of proliferating neoblasts appear to increase at 2-3 days of RNAI but then it goes down to the same level as in controls, whereas there is an increase in the expression of post-mitotic markers. On the other hand there is a significant decrease in apoptotic cell death which the authors interpret as mmp-1 being a positive regulator of apoptosis in planarians. In the case of mt-mmpA its silencing in intact non-regenerating animals also leads them to die. However, here, neoblast proliferation is not affected at any time-point, the expression of post-mitotic markers is reduced and authophagy is significantly increased. From BrdU labeling the authors conclude that mt-mmpA may mediate cell migration during homeostasis.
Further experiments are required to better characterize how the silencing of these MMPs alters the ECM as well as any ECM-cell interaction to explain the severe phenotypes observed. It will be also important to determine which cell types (neoblasts or differentiated cells or both) are involved and why these defects do not appear during regeneration. Still, this work may represent a stimulating starting point to characterize the function of ECM during planarian regeneration and homeostasis.
By combining genetic and optic tools, optogenetics allows to control the activity of individual neurons in vivo. Optogenetics (http://optogenetics.weebly.com/index.html) was developed after the discovery of light activated cation channels and chloride pumps such as Channelrhopdopsin, Halorhodopsin and Archaerhodopsin. One of the big benefits of optogenetics is the spatial and temporal precision with which specific neurons can be selectively targeted. The strategy is to make first a construct in which the gene encoding one if this light-sensitive ion channel is put under the control of a promoter to drive its expression in the desired cells. Then you introduce this construct into, for example, a mouse’s brain or cultured cells (i.e. with the help of a virus) so the light-sensitive ion channel will be expressed in your target cells. Then, by using specific “hardware” (electrodes and fiber-optic cables) you can deliver specific light pulses to your targeted neurons in order to turn them on or off by modulating the activity of such channels. Being able to modulate specific neurons and neuronal circuits can be extremely useful to address some neurological disorders.
Keeping that in mind the laboratory of Michael Levin is using optogenetics to further study and characterize the known endogenous bioelectrical signals that are required to regulate a proper regenerative response in several animals. Thus, for example, during planarian regeneration the anterior blastema is normally depolarized by the action of endogenous H,K-ATPse activity and that triggers head regeneration. By blocking this activity, amazingly, there is a relative hyperpolarization of the blastema and that results in blocking head regeneration (http://www.ncbi.nlm.nih.gov/pubmed/21276941). Xenopus tadpoles are able to regenerate their tails. After amputation a bioelectrical hyperpolarizing signal is required to initiate tail regeneration. But Xenopus tadpoles go through a “refractory” period in which regeneration is not permitted and is accompanied by the absence of an endogenous H+-efflux. Now, in a recent paper by Adams and Tseng, the Levin lab reports the use of Archaerhodopsin-3, a light-gated proton pump to trigger tail regeneration at the “refractory” stage (http://bio.biologists.org/content/early/2013/01/11/bio.20133665.full). Through light activation of an Arch-mediated H+-efflux after tail regeneration at the “refractory” stage, the authors show a significant increase in the regenerative abilities compared to non-light-stimulated controls. Moreover this induced regenerative program activates endogenous signaling programs normally required for regeneration as those regulated by Notch1 and Msx1. The exciting consequences that these results and their further application may have in the field of regenerative medicine are obvious.