(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Pax6 limits the competence of developing cerebral cortical cells to respond to inductive intercellular signals [1]

['Martine Manuel', 'Simons Initiative For The Developing Brain', 'Patrick Wild Centre', 'University Of Edinburgh', 'Edinburgh', 'United Kingdom', 'Kai Boon Tan', 'Zrinko Kozic', 'Michael Molinek', 'Tiago Sena Marcos']

Date: 2022-09

The development of stable specialized cell types in multicellular organisms relies on mechanisms controlling inductive intercellular signals and the competence of cells to respond to such signals. In developing cerebral cortex, progenitors generate only glutamatergic excitatory neurons despite being exposed to signals with the potential to initiate the production of other neuronal types, suggesting that their competence is limited. Here, we tested the hypothesis that this limitation is due to their expression of transcription factor Pax6. We used bulk and single-cell RNAseq to show that conditional cortex-specific Pax6 deletion from the onset of cortical neurogenesis allowed some progenitors to generate abnormal lineages resembling those normally found outside the cortex. Analysis of selected gene expression showed that the changes occurred in specific spatiotemporal patterns. We then compared the responses of control and Pax6-deleted cortical cells to in vivo and in vitro manipulations of extracellular signals. We found that Pax6 loss increased cortical progenitors’ competence to generate inappropriate lineages in response to extracellular factors normally present in developing cortex, including the morphogens Shh and Bmp4. Regional variation in the levels of these factors could explain spatiotemporal patterns of fate change following Pax6 deletion in vivo. We propose that Pax6’s main role in developing cortical cells is to minimize the risk of their development being derailed by the potential side effects of morphogens engaged contemporaneously in other essential functions.

Funding: This research was funded by grants from the Medical Research Council UK (Mr/J003662/1, D.J.P and J.O.M.; Mr/N012291/1, D.J.P.), the Biotechnology and Biological Sciences Research Council UK (Bb/N006542/1, D.J.P. and J.O.M.), the Muir Maxwell Epilepsy Centre ( http://www.muirmaxwellcentre.com/ ; M.I.D.) and Simons Initiative for the Developing Brain ( https://sidb.org.uk/ ; grant number 529085, J.O.M. and D.J.P.), and Principal’s Career Development and Edinburgh Global Research Scholarships from the University of Edinburgh ( https://www.ed.ac.uk/institute-academic-development/postgraduate/doctoral/career-management/principals-scholarships ; K.B.T.) and a Scholarship from the Ministry of Higher Education, Malaysia ( https://www.mohe.gov.my/en ; M.F.A.R.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2022 Manuel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

We began by examining the effects of inducing cortex-specific Pax6 loss-of-function in cortical progenitors using population and single-cell transcriptomics followed by expression analysis of selected genes in tissue sections. The response was dichotomous: Many Pax6-null progenitors continued to generate excitatory neurons that made cortical layers relatively normally, while others adopted abnormal developmental trajectories, the nature of which varied with age and cortical location. Subsequent in vivo and in vitro experiments revealed that Pax6 blocks the deviant trajectories by reducing the ability of cortical cells to react abnormally to substances normally present—and carrying out other essential functions—around them. We propose that the main function of Pax6 in cortical development is to imbue the process with stability and reproducibility by protecting it from potentially destabilizing signals in the cortical environment.

The Pax6 gene emerged 500 to 700 million years ago and has been conserved through all triproblastic animal lineages, where it is involved in many neural and nonneural processes [ 13 , 14 ]. Its expression in the developing brain of extant vertebrates and invertebrates indicates that it acquired important functions very early in this organ’s evolution. In mammalian embryos, it is activated prior to neural tube closure in the anterior neuroectoderm where brain forms [ 15 ]. Its importance for the production of cortical excitatory neurons is demonstrated by the phenotypes of constitutively mutant mouse embryos unable to make functional Pax6. These embryos show reduced cortical expression of genes involved in excitatory neuron production and increased cortical expression of genes involved in the development of subcortically derived cell types including inhibitory interneurons [ 16 – 24 ]. We set out to discover what Pax6 does in cortical progenitors to help govern their normal production of excitatory neurons.

The cerebral cortex is a complex amalgamation of 2 major neuronal cell classes generated by developmental cell lineages expressing different sets of transcription factors [ 6 – 9 ]. One cell class uses the excitatory neurotransmitter glutamate to propagate neuronal activity through cortical circuits and is produced by progenitors located in the developing cerebral cortex itself. It develops from cell lineages that express transcription factors including Pax6, Neurog2, and Eomes. The other cell type uses the inhibitory neurotransmitter gamma aminobutyric acid (GABA) to refine and elaborate patterns of cortical neuronal activity and is produced by progenitors located subcortically. It develops from cell lineages that express substantially different sets of transcription factors. Pax6 is one of the first transcription factors to be expressed differentially between the progenitors of excitatory and inhibitory cortical neurons [ 10 – 12 ], making it a good candidate to be involved in regulating the likelihood of cortical progenitors adopting an excitatory neuronal fate.

Gene regulatory networks (GRNs) modulated by intercellular signals control the generation of the specialized cell types that compose multicellular organisms [ 1 , 2 ]. These control mechanisms affect the developmental trajectories of cells in a variety of ways to guide the production of particular cell types and prevent the emergence of alternatives. Transcription factors whose levels vary among developing cells in precise, reproducible spatiotemporal patterns are essential components of GRNs. In some cases, their regional activation in response to inductive signals drives the production of region-specific cell types, but there are many other ways in which they can operate. For example, they can determine whether, and if so how, cells respond when confronted by inductive signals, i.e., their competence [ 3 , 4 ]. Restricting the competence of cells as they develop is likely to maximize the probability of them following reproducibly their stereotypical developmental trajectories, e.g., by mitigating the effects of biochemical noise in the signals they encounter or in the intracellular pathways processing those signals [ 5 ] and by preventing them responding in inappropriate ways to signaling molecules surrounding them.

Results

Removal of Pax6 from the progenitors of cortical neurons Most cortical excitatory neurons are generated between embryonic day 12.5 (E12.5) and E16.5 in mice [25–29]. They are derived from cortical radial glial progenitors (RGPs), some directly and others indirectly via the initial production of transit-amplifying intermediate progenitors (IPs) [30–32]. All RGPs express Pax6 [11]. We used the Emx1-CreERT2 allele [33] to make tamoxifen-induced cortex-specific homozygous Pax6 conditional knockouts (Pax6 cKOs) (S1A Fig). Heterozygous littermates with deletion in just one Pax6 allele served as controls; previous work on heterozygotes detected no abnormalities in cortical levels and patterns of Pax6 protein expression or cortical morphogenesis, almost certainly because known feedback mechanisms caused compensatory increases in Pax6 production from the normal allele [34–36]. When we gave tamoxifen at E9.5 (tamoxifenE9.5), levels of normal Pax6 mRNA in Pax6 cKOs fell to <50% of control by E11.5, to approximately 10% of control by E12.5 and to almost zero by E13.5 (S1B Fig) and levels of Pax6 protein fell to approximately 5% of control by E12.5 (S1C and S1D Fig). By E12.5, Pax6 was undetectable by immunohistochemistry in almost all RGPs (except those in a narrow ventral pallial domain where Emx1 is not expressed) (S1E and S1F Fig) while a Cre reporter, RCEEGFP (S1A Fig; [37]), was active in most cortical cells (S1E Fig). Thus, tamoxifenE9.5 ensured that the vast majority of cortical neurons was generated, directly or indirectly, from RGPs that had lost Pax6 protein.

Pax6 loss caused ectopic gene expression in cortical cells We first used bulk RNAseq to study the effects of tamoxifenE9.5-induced Pax6 cKO in rostral and caudal cortex at E12.5 and E13.5 (S2A Fig). Raw data are available at the European Nucleotide Archive accession numbers PRJEB5857 and PRJEB6774. We used 4 biological replicates for each location, age and genotype; principal component analysis (PCA) on all datasets taken together showed high-level clustering by age and location (S1G Fig). The number of genes with significantly altered expression levels (adjusted p < 0.05) in Pax6 cKO cortex increased approximately 3-fold between E12.5 and E13.5 (S2B Fig and S1 Table). At each age, the numbers of up-regulated and down-regulated genes were similar. We identified regulated genes with nearby Pax6 binding sites using published chromatin immunoprecipitation-sequencing data from E12.5 forebrain obtained by Sun and colleagues [38]. We followed their assignation of peaks to the gene with the nearest transcription start site (TSS), provided the peak lay within the genomic interval between 50 kb upstream of the TSS and 50 kb downstream of the transcription end site. The proportion of regulated genes with a nearby binding site was higher at E12.5 than E13.5 (S2C Fig), suggesting an accumulation of indirect gene expression changes with age. We then examined which genes altered their expression levels in Pax6 cKO cortex (S2D and S2E Fig). We found that a major effect was the ectopic activation of genes normally expressed only extracortically, either by surrounding noncortical telencephalic cells or by cells normally located outside the telencephalon (S2D and S2E Fig). Many of these genes encoded transcription factors known to be involved in cell specification [16,17,24,39,40–55]. Note that our study did not aim to provide new evidence on whether genes with altered cortical expression were normally directly regulated by Pax6 binding to their enhancers or promoters (for previous data on this, in addition to those used above in [38], see [19,56,57]). In summary, these findings indicated that acute conditional cortex-specific Pax6 removal rapidly affected the specification of at least some embryonic cortical cells.

Pax6 loss induced eGC production in a distinct spatiotemporal pattern We next examined cells that deviated to the eGC fate by probing for expression of Gsx2, Dlx1, and Gad1 (S9A–S9D Fig). In normal cortical development, Gsx2 becomes active only in small numbers of late-stage (E16.5 or older) cortical SVZ cells that generate cell types other than cortical neurons [89] (these cells are seen in S9A Fig: “Control E16.5”). Following tamoxifenE9.5, a wave of ectopic Gsx2 activation was advancing rapidly across the cortex by E12.5. It began laterally and swept progressively further medially to occupy all parts of lateral cortex by E14.5 but did not extend all the way through medial cortex (Figs 3H and S9A). We examined the extent to which this change depended on when tamoxifen was administered (evidence in S10A and S10B Fig confirmed that tamoxifen administration at ages other than E9.5 also caused Pax6 removal from most RGPs within 3 d). We found similar distributions of Gsx2+ cells at E13.5 no matter whether tamoxifen was administered on E8.5, E9.5, or E10.5 (Figs 3H and S9A) and even in E13.5 constitutive Pax6−/− mutants that had never expressed functional Pax6 (S9A Fig). When we administered tamoxifen later, on E13.5, Gsx2+ cells were distributed throughout the entire lateral cortex 3 d later. This resembled the distributions at similarly late ages (E14.5 to E16.5) following early tamoxifen administration (E8.5 to E10.5) and not the distributions 3 d after early tamoxifen administration (S9A Fig). We concluded that the spatial distribution of Gsx2+ cells depended mainly on cortical age rather than time elapsed since Pax6 removal, suggesting that cortical factors that change with age have important influences on the outcome of Pax6 removal. TamoxifenE9.5 induced a wave of ectopic Dlx1 expression similar to that of Gsx2 expression, i.e., it was underway by E12.5 (S9B Fig) and had spread through lateral cortex but only encroached to a limited extent into medial cortex by E14.5 (Fig 3H). TamoxifenE9.5 also led to the generation of a large population of Gad1+ cells in the lateral cortex (S9C and S9D Fig). Most of these cells were cortically derived (i.e., they were GFP+ Emx1-lineage) but they were intermingled throughout their domain with other Gad1+ cells that were GFP-negative subcortically generated immigrants (arrows in S9D Fig). In the VZ and SVZ of Pax6-deleted lateral cortex, Gsx2, Dlx1, and Gad1 were activated by partially overlapping bands of cells centered progressively further basal to the ventricular surface (S9E Fig). The Gsx2+ and Dlx1+ bands overlapped the basal side of the Ascl1+ band and the Dlx1+ band extended further basally than the Gsx2+ band. This was followed by the Gad1+ band, which showed considerable overlap with the Dlx1+ band but less overlap with the Ascl1+ and Gsx2+ bands (summarized in S9F Fig). Where domains of expression overlapped, coexpressing cells were frequent. Small proportions of Gsx2+ or Dlx1+ cells coexpressed Eomes (arrows in S9G Fig), in agreement with findings in our scRNAseq data (Figs 2C and S5C). We concluded that the production of eGCs unfolded in a distinct spatiotemporal pattern in mainly lateral cortex.

Pax6 loss induced ectopic Olig2 expression largely independently of eGC production We then examined the pattern of ectopic cortical activation of Olig2, which is expressed in progenitors that generate cortical interneurons and oligodendrocytes, is normally restricted to the embryonic subpallium at around E13.5 (S9H Fig: “Control E13.5”) and later spreads as Olig2+ cells migrate into the cortex (S9H Fig: “Control E16.5”) [33,90]. Our scRNA-seq data indicated that Olig2 was not specifically marking eGCs but was expressed by many additional cell types including RGPs, aPs, IPs, and differentiating cells in Pax6 cKOs (S6 and S10C Figs). Its ectopic spatiotemporal activation pattern differed from that of Gsx2, Dlx1, and Gad1 to the extent that it appeared throughout the entire lateral cortex earlier, by E13.5, but was similar in showing relatively little activation in medial cortex, even at later ages (S9H Fig). The domain of Olig2 activation was similar in E13.5 to E16.5 embryos regardless of whether tamoxifen was given at E9.5, E10.5, or E13.5. In lateral cortex, many progenitors coexpressed Olig2 and Ascl1 (S9I Fig); this was supported by scRNAseq data showing that 51.8% and 67.3% of Olig2+ cells expressed Ascl1 at E13.5 and E14.5, respectively. Nevertheless, our Pax6 cKO E14.5 scRNAseq data detected Olig2 coexpression in only a small proportion (9.6%) of cells expressing eGC markers Gsx2, Dlx1, and Gad1 (S10 Fig). These findings suggested that the Pax6-loss-induced activation of Olig2 and of eGC-expressed genes such as Gsx2, Dlx1, and Gad1 occurred largely independently. They provided further evidence of spatiotemporal variation in the effects of Pax6 loss on the ectopic activation of different genes.

The eGCs were highly proliferative Our scRNAseq data indicated the existence of a substantial population of proliferating eGCs in E14.5 Pax6 cKO cortex. This was demonstrated, for example, by the rising levels of the mitotic marker Mki67 along the inferred pseudotime trajectory of the lineage leading to eGC-P generation (Fig 4A; trajectories were obtained using Slingshot and tradeSeq; [91–93]). To test this conclusion further, we used the Emx1-CreERT2 allele with tamoxifenE9.5 to delete Pax6 and then labeled proliferating cells by administering the S phase marker 5-ethynyl-2′-deoxyuridine (EdU) at E13.5, 30 min before death (Fig 4B). We reacted sections for EdU and Gsx2, a marker of early eGCs (and also for GFP from a Btg2-GFP transgene that was incorporated into the mice for reasons given below) (Fig 4C). Most Gsx2+ cells were in S phase (mean = 59.0% ± 3.4 SD; counts were from 20 equally spaced coronal sections through the cortex for each embryo; n = 5 embryos from separate litters; Sheet A in S3 Data), confirming their high level of proliferation. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. The proliferation, coalescence, and electrophysiological properties of Gsx2-lineage eGCs. (A) Major pseudotemporal trajectories inferred from E14.5 Pax6 cKO scRNAseq data (one leading to eGC-Ps, one to eGC-Ns, and one to cortical glutamatergic neurons) and expression of the marker of proliferating cells, Mki67, along each. (B) The experimental procedure for EdU labeling. The Emx1-CreERT2 allele with tamoxifenE9.5 was used to delete Pax6 (embryos carried a Btg2-GFP transgene); EdU was given at E13.5, 30 min before death; 20 coronal sections equally spaced through the brain were immunoreacted for EdU, Gsx2, and GFP; counts were made in the boxed area. (C) Fluorescence quadruple-staining for Gsx2, EdU, GFP (marking Btg2-expressing cells), and DAPI in E13.5 Pax6 cKO cortex after the procedure in (B). Scale bar: 0.1 mm. (D) Fluorescence double-staining for Sox9 and GFP (marking Btg2-expressing cells) in E13.5 Pax6 cKO cortex after tamoxifenE9.5. Scale bar: 0.1 mm. (E) Fluorescence double-staining for Gsx2 and GFP (marking Btg2-expressing cells) in E13.5 and E14.5 Pax6 cKO cortex after tamoxifenE9.5. Scale bar: 0.1 mm. (F) Eomes immunoreactivity and methyl green counterstaining in control and E16.5 Pax6 cKO cortex after tamoxifenE9.5. Scale bar: 0.1 mm. (G) Fluorescence immunoreactivity for GFP (Emx1-lineage) and in situ hybridization for Gad1+ cells in E18.5 Pax6 cKO cortex after tamoxifenE9.5. Scale bars: 0.1 mm and 0.01 mm. (H) Quantifications of the total numbers of Gad1+ cells in the lateral CP (red: they were GFP-, subcortically derived), of Gad1− cells in the CP (green: GFP+, cortical-born) and of cells in the Pax6 cKO sub-CP masses (pink: Gad1+, GFP+) in control and Pax6 cKO E18.5 embryos after tamoxifen at E9.5 (for quantification method, see S11F Fig). Total numbers of cells were greater in Pax6 cKO cortex (p < 0.05) and numbers of lateral CP cells were reduced (p < 0.02) (averages ± SEM; Student paired t tests; n = 4 embryos of each genotype, from 4 independent litters) (S4 Data). (I) The experimental procedure for electrophysiology (J-P). The Emx1-Cre allele was used to delete Pax6; embryos carried a GFP reporter transgene. Recordings were from sub-CP masses at P3–10. (J) Sub-CP mass in P7 slice prepared for electrophysiology: The cortex was GFP+ and the sub-CP mass was intensely so. Scale bar: 0.5 mm. (K,L) Examples of responses of sub-CP mass cells to current injections (square steps, magnitudes color-coded, 500 ms duration). Membrane voltages were held at −70mV. Some cells produced small spikelets (J), others did not (K). (M) TTX (300 nM) reduced spikelet amplitudes; examples of entire response and spikelet alone before and after TTX application; effects of TTX were significant (p = 0.035, Wilcoxon signed rank test; n = 6 cells) (Sheet A in S5 Data). (N-P) Passive electrical properties of P3-P10 sub-CP mass cells compared to P5-P7 cortical cells from layer 5 of somatosensory area 1 (n = 66 sub-CP mass cells, Sheet B in S5 Data; n = 49 cortex cells, data for these CP cells are in S5 Table). Sub-CP mass cells had significantly lower capacitance (p = 2.2 × 10−16, Mann–Whitney test) and significantly higher input resistance (p = 2 × 10−10, Mann–Whitney test) and resting membrane potential (p = 2.2 × 10−16, Mann–Whitney test). For capacitance, values were significantly higher among sub-CP mass cells that produced spikelets (n = 22 cells; n = 44 produced no spikelet) (p = 1.9 × 10−9, Mann–Whitney test). CP, cortical plate; EdU, 5-ethynyl-2′-deoxyuridine; GFP, green fluorescent protein; Pax6 cKO, Pax6 conditional knockout; TTX, tetrodotoxin. https://doi.org/10.1371/journal.pbio.3001563.g004 We then studied the types of division that Pax6 cKO cortical progenitors made. Previous work has shown that RGPs (Sox9+) and IPs (Eomes+) produce either postmitotic neurons or new progenitors [30,31,94,95]. Progenitors of the latter type, often described as proliferative progenitors, do not express the antiproliferative gene Btg2; others, often described as neurogenic, do express Btg2 [96,97]. We used the Btg2-GFP transgene [97] with immunohistochemistry to identify neurogenic progenitors (Fig 4D and 4E). Many Gsx2+ cells expressed Btg2 at E13.5 and E14.5, but a sizeable minority did not. Quantification in E13.5 tissue sections showed that 68.1% ± 6.5 (SD) of Gsx2 protein-expressing cells were also Btg2-expressing (counts were made in 20 equally spaced coronal sections through the cortex for each embryo; n = 5 embryos from separate litters; Sheet A in S3 Data). This was similar to scRNAseq data, which showed Btg2 expression in 76.2% and 74.0% of Gsx2+ cells at E13.5 and E14.5, respectively. These data indicated that, overall, about a quarter of the cortical cells that activated Gsx2 were proliferative (i.e., Btg2-nonexpressing; their daughters would divide at least once more). The emergence in Pax6 cKO cortex of substantial numbers of repeatedly and rapidly dividing progenitors caused a large expansion of the eGC population, described in the next section.

Sub-CP mass cells showed immature electrophysiological properties We tested whether sub-CP mass cells developed electrophysiological properties resembling those of interneurons by making whole-cell current-clamp recordings at P5 to P10 (Fig 4I) [98–100]. These ages encompassed those by which normal cortical neurons have acquired the ability to generate individual or trains of action potentials (APs) in response to depolarizing stimuli [101–103]. The sub-CP masses were easily identified in slices at all ages by their intense GFP expression (Fig 4J). The properties of the sub-CP mass cells were similar across the range of ages studied here. None of them generated mature APs. A third (22/66) produced either spikelets (spikelet peak < 10 mV; spikelet amplitude = 5 to 25 mV; little or no afterhyperpolarization (AHP); Fig 4K) or, in 2 cases, underdeveloped APs (peak amplitude > 30 mV and AHP > 15 mV). Most (44/66) produced neither (Fig 4L). Spikelet amplitudes were reduced by approximately 90% following the addition of 300 nM tetrodotoxin (TTX), which blocks the voltage-gated Na+ channels responsible for the rising phase of the AP [104,105] (Fig 4M), suggesting that spikelets were immature APs. One possibility was that the cells that produced spikelets were eGC-Ns, whereas those that did not were eGC-Ps. The sub-CP mass cells had much lower capacitances and higher input resistances (R in s) and resting membrane potentials (RMPs) than P5 to P7 cortical neurons recorded in layer 5 of primary somatosensory cortex (Figs 4N–4P and S14D–S14F and S5 Table). Their relatively low capacitances were a sign that they had relatively small somas (Fig 4N). When we split them into those that produced spikelets and those that did not, we found that the former had higher capacitances, indicating that they were slightly larger (Fig 4N). The relatively high R in s and RMPs of the sub-CP mass cells, neither of which differed significantly between cells that did or did not generate spikelets, were likely attributable to immaturity in the numbers of ion channels in their cell membranes [103,106–108]. We concluded that although the transcriptomes of these cells showed progress toward a GABAergic interneuron fate, they were unable to develop corresponding cellular properties. Whether this was because they had a cell autonomous inability to mature and/or a problem with the environment in which they found themselves was not tested here.

Why Pax6 deletion altered the fates of only some cortical cells: A hypothesis We then turned to the question of why some cortical cells switched fate while others did not after Pax6 deletion from cortical progenitors. A parsimonious explanation was that Pax6 loss increased the potential for all RGPs to generate inappropriate cell lineages, but triggering this required additional, extracellular factors. Systematic cross-cortical variations in the types and levels of these factors might have been responsible for generating the spatiotemporal patterns of normal and abnormal specification seen after Pax6 deletion. We set out to test this idea.

Immigrating cortical interneurons enhanced the misspecification of Pax6 cKO cortical cells The striking similarity between the spatiotemporal characteristics of the wave of eGC production and the wave of subcortically generated interneuron immigration (S15A Fig), which was not disrupted by Pax6 removal (see above), suggested that the immigrating interneurons might have been one source of extracellular factors triggering abnormal specification among Pax6 cKO cortical cells. To test this possibility, we removed subcortical tissue from one side of cultured coronal slices of E13.5 Pax6 cKO (tamoxifenE9.5) telencephalon to prevent further interneuron influx and compared the production of Gsx2+ cells on the 2 sides after 48 h in culture, using the GFP reporter to mark cells of cortical origin (Fig 6A and 6B). The numbers of subcortically generated interneurons (i.e., GFP-negative Gad1+ cells) were approximately 4 times higher on the intact side (Fig 6C and 6D), as anticipated from previous work using this approach [6]. Proportions of GFP+ Gsx2+ cells were several times higher on the intact side, with significant differences in the more lateral parts of cortex (Fig 6E and 6F). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. Extracellular signals promoted eGC production in Pax6 cKOs cortex. (A) The experimental procedure for (B-F): TAM was given at E9.5 to generate Pax6 cKOs, with Cre-deleted cells expressing GFP; coronal slices were cultured on E13.5 with the ventral telencephalon removed on one side; after 2 DIV, sections from cultured slices were cut and processed. Gsx2+ GFP+ cells were counted in 3 ROIs on each side. (B, C) GFP immunoreactivity and in situ hybridizations for Gad1 in sections prepared as in (A). Scale bars: 0.1 mm and 0.01 mm. (D) Average (±SEM) numbers of immigrant Gad1+ interneurons (i.e., GFP nonexpressing) per section were lower on the side lacking ventral telencephalon (n = 3 independent cultures; Student paired t test) (Sheet A in S6 Data). (E) Gsx2 immunoreactivity in sections prepared as in (A). Scale bar: 0.1 mm. (F) Average (±SEM) proportions of GFP+ cells that were Gsx2+ in each ROI in (A) (n = 3 independent cultures; Student paired t tests; n.s., not significant) (Sheet B in S6 Data). (G) Immunoreactvity for Shh in control telencephalic sections at E13.5 and E15.5 (see S15B Fig for evidence of antibody specificity). Scale bar: 0.1 mm. (H) The experimental procedure for (I, J): Vismodegib or vehicle alone was injected into the ventricle of E14.5 Pax6 cKO embryos made using Emx1-Cre; central regions of lateral cortex from coronal sections at 3 rostral-to-caudal levels were analysed at E15.5. (I) Gsx2 immunoreactivity in boxed region in (H). Scale bar: 0.1 mm. (J) Average (±SEM) proportions of cells in the PZs and CPMs that were Gsx2+ (n = 5 embryos from 3 litters given vehicle alone; n = 6 embryos from 3 litters given vismodegib; Student t tests) (Sheet A in S7 Data). (K) The experimental procedure for (L, M): Constructs expressing Smo shRNA + GFP or scrambled shRNA + GFP were electroporated into the cortex of E14.5 Pax6 cKO embryos made using Emx1-Cre; electroporated cells were analysed at E15.5 (as in S15D Fig). (L) Gsx2 and GFP immunoreactivity in electroporated regions. Scale bar: 0.01 mm. (M) Cumulative frequency distributions of the intensity of Gsx2 immunoreactivity in electroporated cells (GFP+; green) and surrounding randomly selected non-electroporated cells (GFP−; black) for the 2 constructs (see Figs 6K and S15D) (n = 3 embryos from 3 litters given Smo shRNA; n = 4 embryos from 3 litters given scrambled shRNA; Kolmogorov–Smirnov tests) (Sheet B in S7 Data). (N) The experimental procedure for (O): TAM was given at E9.5 to generate Pax6 cKOs and Cre-deleted cells expressed GFP; coronal slices of telencephalon were cultured on E13.5; cyclopamine or vehicle alone were added either on beads or in solution (10 μM); slices were cultured for 2 DIV. (O) Sections from cultured slices obtained as in (N) were immunoreacted for Gsx2 and GFP. Scale bar: 0.1 mm. Ctx, cortex; DIV, day in vitro; GFP, green fluorescent protein; Hem, cortical hem; LGE, lateral ganglionic eminence; Pax6 cKO, Pax6 conditional knockout; PSPB, pallial–subpallial boundary; PZ, proliferative zone; ROI, region of interest; sub-CPM, sub-cortical plate mass; TAM, tamoxifen. https://doi.org/10.1371/journal.pbio.3001563.g006 This outcome suggested that the proportions of Pax6 cKO cortical cells that deviated to develop as eGCs was influenced by extracellular factors.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3001563

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.