Neuromesodermal progenitors and the making of the spinal cord

© 2015. Published by The Company of Biologists Ltd | Development (2015) 142, 2864-2875 doi:10.1242/dev.119768 HYPOTHESIS Neuromesodermal progenitors and the making of the spinal cord ABSTRACT Neuromesodermal progenitors (NMps) contribute to both the elongating spinal cord and the adjacent paraxial mesoderm. It has been assumed that these cells arise as a result of patterning of the anterior neural plate. However, as the molecular mechanisms that specify NMps in vivo are uncovered, and as protocols for generating these bipotent cells from mouse and human pluripotent stem cells in vitro are established, the emerging data suggest that this view needs to be revised. Here, we review the characteristics, regulation, in vitro derivation and in vivo induction of NMps. We propose that these cells arise within primitive streak-associated epiblast via a mechanism that is separable from that which establishes neural fate in the anterior epiblast. We thus argue for the existence of two distinct routes for making central nervous system progenitors. KEY WORDS: Neuromesodermal progenitors, Wnt, FGF, Bipotent cells, Neural induction, Spinal cord, Stem cells Introduction The vertebrate central nervous system (CNS) is first manifest as an ovoid region of thickened epiblast cells in front of the organiser/ anterior primitive streak. This region is known as the anterior neural plate (Fig. 1). Fate-mapping studies in a range of vertebrate species all show that the forebrain forms in the rostralmost part of this region, whereas more posterior regions of the CNS (midbrain and hindbrain) arise from cells positioned closer to the primitive streak. The position of the prospective hindbrain/spinal cord is more variable between species; in the chick, for example, this is located closest to the primitive streak (Spratt, 1952), whereas in the mouse embryo some laterally positioned epiblast cells also move medially to contribute to posterior neural tissue (Lawson and Pedersen, 1992). The prevailing view of vertebrate neural induction derives largely from work in the amphibian embryo. This proposes that initial induction of the anterior neural plate is followed by the formation of more posterior neural regions via patterning of this anterior tissue with posteriorising signals (to form posterior neural plate) (Fig. 2A). This view was first formulated in the so called ‘activationtransformation’ hypothesis proposed by Nieuwkoop (Nieuwkoop, 1952; Nieuwkoop and Nigtevecht, 1954), in which ‘activation’ involved the induction of anterior neural tissue and ‘transformation’ implied its patterning to more posterior character (Fig. 2A). This was subsequently substantiated at the molecular level with the discovery that inhibition of bone morphogenetic protein (BMP) 1 Instituto de Medicina Molecular and Instituto de Histologia e Biologia do Desenvolvimento, Faculdade de Medicina da Universidade de Lisboa, Avenida 2 Prof. Egas Moniz, Lisboa 1649-028, Portugal. Division of Cell & Developmental Biology, College of Life Sciences, University of Dundee, Dow Street, Dundee DD1 5EH, UK. *Author for correspondence () This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. 2864 signalling promoted the formation of anterior neural tissue (with forebrain character), which could then be patterned by posteriorising signals, such as retinoic acid (RA), Wnt and fibroblast growth factors (FGFs). The molecular basis for this ‘activation’ step is not without controversy when extended to amniote embryos. Although inhibition of BMP signalling promotes neural fate in the mouse embryo, for example (Di-Gregorio et al., 2007), BMP inhibition alone is insufficient to induce neural tissue in the chick extraembryonic epiblast (Stern, 2006). This might reflect differences in experimental assays, especially the timing of manipulations, and/or the operation of species-specific mechanisms. It is also now recognised that neural induction is a complex multistep process. This includes roles for FGF signalling as the mediator of an early unstable ‘preneural’ state in the chick embryo, which is then stabilised by further (yet to be identified) signals (Stern et al., 2006). However, it should be noted that some studies have not found a requirement for FGF/Erk signalling during neural differentiation, for example in embryonic stem cells (ESCs) and epiblast-derived stem cells (EpiSCs) (Greber et al., 2010, 2011; Ozair et al., 2013b; Hamilton and Brickman, 2014). Wnt signalling, or its antagonism, is also variably implicated in this ‘activation’ step in different species. Wnt, FGF and RA signalling then subsequently act as local posteriorising factors, while Wnt antagonism promotes anterior/ forebrain identity. Detailed reviews of neural induction are provided elsewhere (Stern, 2005, 2006; Ozair et al., 2013a; Andoniadou and Martinez-Barbera, 2013). However, a common premise here is that the acquisition of neural fate starts with induction of the anterior neural plate, and that this is achieved as a result of events in the anterior epiblast, which gives rise to the entire CNS. The discovery of a bipotent neuromesodermal progenitor (NMp) that contributes to both the spinal cord and paraxial mesoderm in the mouse embryo (Tzouanacou et al., 2009) has now raised the possibility that some posterior neural tissue is generated independently of the mechanism(s) that induces the anterior neural plate. The idea that the posterior spinal cord arises from progenitor cells with a neuromesodermal potential was proposed as long ago as 1884, based on morphological observations (Kölliker, 1884), and there has been a long-running debate about whether head, trunk and tail regions of vertebrate embryos are induced by distinct mechanisms (Handrigan, 2003; Stern et al., 2006). In more recent years, fate-mapping studies of groups of cells in mouse and chick embryos at late primitive streak to tailbud stages (Brown and Storey, 2000; Iimura and Pourquié, 2006; Cambray and Wilson, 2007; Olivera-Martinez et al., 2012) have localised this NMp cell population to the caudal lateral epiblast (CLE; also known as the stem zone or caudal neural plate in chick) and adjacent node-streak border (NSB) (Fig. 1). Recent studies have also demonstrated that mouse ESCs and EpiSCs, as well as human ESCs, can be directed to form NMps in vitro (Gouti et al., 2014; Tsakiridis et al., 2014; Turner et al., 2014a; Denham et al., 2015; Lippmann et al., 2015; Tsakiridis and Wilson, 2015), raising the possibility of exploring the potential therapeutic use of NMps (see Box 1). These cells can be DEVELOPMENT Domingos Henrique1, Elsa Abranches1, Laure Verrier2 and Kate G. Storey2, * HYPOTHESIS Development (2015) 142, 2864-2875 doi:10.1242/dev.119768 A E7.5 mouse embryo MB (...truncated)