Compared to other regeneration models, freshwater planarians display a distinctive feature: they regenerate from a unique population of adult pluripotent stem cells, called neoblasts. This makes planarian an excellent model in which to study the behaviour of stem cells in vivo. Although classically seen as a rather homogeneous cell population, several recent studies have suggested that neoblasts could, in fact, constitute a wide heterogeneous population, based on different parameters and features, in which, for instance, several progenitor-like cells for different cell types exist. Now, a beautiful paper from the laboratory of Peter Reddien goes much deeper into their analysis of planarian stem cells and identifies two major and functionally distinct cellular compartments among neoblasts (http://www.ncbi.nlm.nih.gov/pubmed/25017721).

In this study the authors characterized the neoblasts at a single-cell resolution level. Neoblasts are the only proliferative cells in planarians. Fluorescence-activated cell sorting (FACS) can be used to isolate proliferating neoblasts based on their DNA content. The authors then analysed the expression of 96 genes within each of hundreds of individual neoblasts. These genes were selected from a neoblast transcriptome and included, among others, well-known neoblasts and post-mitotic markers, as well as a variety of transcription factors and regulators highly abundant in neoblasts. Hierarchical clustering allowed distinguishing two main classes of equally sized populations of neoblasts: the zeta-class and the sigma-class. The zeta-class neoblasts were characterized by a high expression of zfp-1, g6pd, fgfr-1, p53, soxP-3 and egr-1. On the other hand, the sigma-class neoblast showed low expression levels of the previous genes and high expression of a distinct set of genes: soxP-1, soxP-2, soxB-1, smad6/7, inx-13, pbx-1, fgfr-4 and nlk-1.

In order to discard that these two neoblast populations could be in fact a single population but at different state of the cell cycle the authors isolated by FACS different neoblast populations at different stages of the cell cycle according to their DNA content. Single-cell profiling showed that both classes, zeta- and sigma-neoblasts, were equally present throughout the cell cycle. Also, zeta- and sigma-neoblasts showed a similar broad spatial distribution along the planarian body. Next, the authors checked how these two populations responded to either amputation (anterior or posterior regeneration) or sublethal irradiation. In all cases, after two days of amputation or irradiation the relative abundances of the zeta- and sigma-classes were the same as in control, untreated animals.

Upon amputation, neoblasts display a bimodal proliferating response with a first mitotic peak at 6 hours (all throughout the regenerating fragment) and a second peak at 48 hours (localized at the wound region). Remarkably, the sigma-neoblasts were overrepresented among the mitotic population both at 6 and 48 hours, indicating that these two mitotic peaks derived mainly from the activity of the sigma-neoblasts. Moreover, analyses on the spatial distribution of these neoblasts indicated that the accumulation of neoblasts at the wound region at 48 hours of regeneration depended mainly on the sigma-class neoblasts.

Next, the authors focussed on the functional analysis of zfp-1, a gene specific to the zeta-class. Upon its silencing by RNAi, the animals died in few weeks. Zfp-1 RNAi resulted in the loss of expression of other zeta-class markers without affecting the expression of sigma-class genes. In fact, the silencing of zfp-1 eliminated the zeta-class neoblasts without affecting the sigma-class cellular compartment. In terms of function, the depletion of the zeta-class neoblasts did not interfere with the normal behaviour of the sigma-class as these cells could mount a proper regenerative response generating two mitotic peaks at 6 and 48 hours as in controls. As the animals in which zfp-1 was silenced were still capable of regenerating a blastema the authors checked whether specific cell lineages were affected in those animals. Remarkably, they found that brain, gut, muscle, protonephridia, eyes and pharynx tissues were apparently normal, indicating that the sigma-class neoblasts were capable of differentiating into a broad range of cell types. However, other cells characterized by the expression of previously identified as neoblast early and late progeny markers were depleted. These genes included prog-1 (early progeny) and AGAT-1 (late progeny). These genes label subepidermal cells that undergo rapid cell turnover. However, it was not clear whether those cells defined a particular cell lineage. Here, by doing RNAseq, the authors found out that several transcripts associated with epidermis, cilia and secretory cells were reduced after the silencing of zfp-1. Specifically, they identified nine genes expressed subepidermally (as prog-1) and seven genes expressed at the epidermis, and whose expression was clearly affected after zfp-1 RNAi. Consequently, the epidermal cells were disorganized, thinner and less abundant in those animals. Moreover, after two weeks of BrdU incorporation much fewer epidermal cells were labelled with BrdU, suggesting that zeta-class neoblasts gives rise, at least in part, to an epidermal cell lineage, probably through an intermediate stage defined by the expression of prog-1 and AGAT-1.

Finally, and through some elegant experiments of transplantation of sigma-class neoblasts into previously irradiated (and therefore depleted of any neoblast) planarians, the authors showed how the sigma-class neoblast were capable of generating the zeta-class compartment. Then, and by checking the gene expression profile of individual cells at different stages of the cell cycle, the authors suggest that zeta-class neoblast would derive from sigma-class cells just following their entry into S phase.

In summary, this study identifies two clearly distinct classes of neoblasts based on their gene expression profiles, response to wounding and cell differentiation potential. The data presented here also indicates that these two classes can be also rather heterogeneous themselves. Thus, for instance, within the sigma-class the authors suggest the existence of a subclass, the gamma-neoblast, characterized by the expression of some genes that had been previously related to the planarian gut, indicating that those gamma-class cells could be related to the gut lineage.