Clonal analysis reveals laminar fate multipotency and daughter cell apoptosis of mouse cortical intermediate progenitors

Abstract

In developing cerebral cortex, most pyramidal-projection neurons are produced by intermediate progenitors (IPs), derived in turn from radial glial progenitors. Although IPs produce neurons for all cortical layers, it is unknown whether individual IPs produce multiple or single laminar fates, and the potential of IPs for extended proliferation remains uncertain. Previously, we found that, at the population level, early IPs (present during lower-layer neurogenesis) produce lower- and upper-layer neurons, whereas late IPs produce upper-layer neurons only. Here, we employed mosaic analysis with double markers (MADM) in mice to sparsely label early IP clones. Most early IPs produced 1-2 neurons for deep layers only. Less frequently, early IPs produced larger clones (up to 12 neurons) spanning lower and upper layers, or upper layers only. The majority of IP-derived clones (∼66%) were associated with asymmetric cell death after the first division. These data demonstrate that laminar fate is not predetermined, at least in some IPs. Rather, the heterogeneous sizes and laminar fates of early IP clones are correlated with cell division/death/differentiation choices and neuron birthdays, respectively.

Apical progenitors remain multipotent throughout cortical neurogenesis

“…S9a). Upon examination of single integration sites at P7, transplanted IPs gave rise to a smaller number of cells than APs, consistent with a more differentiated status and a limited number of divisions (Gao et al, 2014; Mihalas et al, 2016; Mihalas and Hevner, 2018) (Fig. S9b).…”

“…They are born from dynamic subtypes of progenitors ( i.e . apical and intermediate progenitors) whose molecular identities and neurogenic potential are increasingly understood (Gaspard et al, 2008; Gao et al, 2014; Okamoto et al, 2016; Yuzwa et al, 2017; Mihalas and Hevner, 2018), but whose corresponding fate potential remains unknown. While the aggregate ability of cortical progenitor populations to generate earlier-born neuron types when transplanted in a younger host appears to be lost (Frantz and McConnell, 1996; Desai and McConnell, 2000), whether individual subtypes of progenitors have the potential to revert to previous temporal states remains unexamined.…”

“…In these studies, the fate plasticity identified here may thus have been occluded by the overwhelming predominance of other, potentially less plastic, progenitor cell types, and particularly by IPs. In addition, in contrast to APs, which divide multiple times, IPs are thought to divide only once or twice before reaching a terminal division (Mihalas et al, 2016; Mihalas and Hevner, 2018). As a consequence, radiolabeled thymidine, which was used in these early studies, will be more diluted in the neuronal progeny of APs than in the progeny of IPs, potentially biasing quantifications towards the latter lineage.…”

Direct readout of neural stem cell transgenesis with an integration-coupled gene expression switch

“…We also demonstrate that accurate cell-type specific Cre/lox recombination, an approach until now largely restricted to transgenic animal lines, can be achieved using electroporated iOn transgenes. This enabled us to trace in vivo the lineage of individual retinal progenitors defined by expression of the transcription factor Atoh7, revealing frequent terminal divisions that generate neurons of a same class, reminiscent of those observed with mouse cortical intermediate progenitors or Drosophila ganglion mother cells (Holguera and Desplan, 2018;Mihalas and Hevner, 2018). Such lineage pattern was particularly frequent for photoreceptors and amacrine cells, suggesting that progenitors restricted to generate these two cell types may exist in the chicken retina, in complement of those previously described for horizontal cells (Rompani and Cepko, 2008).…”

A Genome-wide Library of MADM Mice for Single-Cell Genetic Mosaic Analysis

“…The temporally controlled induction of MADM with its dual marker property provides exact and unique information not only about birth dates of clones but also regarding their cell division patterns. Thus, MADM has been frequently utilized in the past to study the proliferation behavior of progenitor stem cells in a variety of tissues -including embryonic and adult neural stem cells (Beattie et al 2017;Bonaguidi et al 2011;Espinosa and Luo 2008;Gao et al 2014;Kaplan et al 2017;Liang et al 2013;Llorca et al 2019;Lv et al 2019;Mayer et al 2015;Mihalas and Hevner 2018;Ortiz-Alvarez et al 2019;Picco et al 2019;Shi et al 2017;Wong et al 2018); cardiomyocyte- (Ali et al 2014;Devine et al 2014;Mohamed et al 2018) and pancreatic progenitor cells (Brennand et al 2007;Desgraz and Herrera 2009;Salpeter et al 2010); progenitors in the developing kidney (Riccio et al 2016); and mesenchymal progenitors in the developing lung (Kumar et al 2014). MADM enables high resolution single cell/clonal labeling and permits tracing of complex morphogenetic processes in 4D by using live-imaging over prolonged periods besides analysis of static time points (Hippenmeyer et al 2010;Riccio et al 2016).…”