Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Sep 2.
Published in final edited form as: Neuron. 2015 Aug 20;87(5):989–998. doi: 10.1016/j.neuron.2015.07.011

Clonally related forebrain interneurons disperse broadly across both, functional areas and structural boundaries

Christian Mayer 1, Xavier H Jaglin 1, Lucy V Cobbs 1, Rachel C Bandler 1, Carmen Streicher 2, Constance L Cepko 3, Simon Hippenmeyer 2, Gord Fishell 1,*
PMCID: PMC4560602  NIHMSID: NIHMS717516  PMID: 26299473

Abstract

The medial ganglionic eminence (MGE) gives rise to the majority of mouse forebrain interneurons. Here, we examine the lineage relationship among MGE-derived interneurons using a replication-defective retroviral library containing a highly diverse set of DNA barcodes. Recovering the barcodes from the mature progeny of infected progenitor cells enabled us to unambiguously determine their respective lineal relationship. We found that clonal dispersion occurs across large areas of the brain and is not restricted by anatomical divisions. As such, sibling interneurons can populate the cortex, hippocampus striatum and globus pallidus. The majority of interneurons appeared to be generated from asymmetric divisions of MGE progenitor cells, followed by symmetric divisions within the subventricular zone. Altogether, our findings uncover that lineage relationships do not appear to determine interneuron allocation to particular regions. As such, it is likely that clonally-related interneurons have considerable flexibility as to the particular forebrain circuits to which they can contribute.

Introduction

Understanding the principles by which brain circuits are constructed is a fundamental goal in developmental neuroscience. The assembly of complex brain circuitry begins with the generation of prescribed cell types from what are thought to be stereotyped lineages. While neuronal lineages have been well documented in invertebrates (Kohwi and Doe, 2013), the relationships between progenitors and their progeny are less clear in vertebrates. A retroviral lineage approach was developed over 25 years ago that provided the means to identify lineages within different areas of the CNS (Golden et al., 1995; Walsh and Cepko, 1992). These efforts provided the first indications that lineages within the cortex do not necessarily result from a precisely orchestrated transposition of cells from the proliferative zone to postmitotic areas.

Following these original studies, a number of methods have been developed that allow for the directed labeling and tracking of progenitors from discrete proliferative zones. These methods have employed the use of transgenic driver lines, which direct the labeling of neuronal clones to particular proliferative zones (Ciceri et al., 2013; García-Moreno et al., 2014; Vasistha et al., 2014; Yu et al., 2009; Zong et al., 2005). In the cortex, the use of these methods indicated that pyramidal cell migration is more radially coherent than previously thought. Indeed, while some degree of dispersion occurs, the majority of clones populate columns and even participate functionally in common circuits as predicted by Rakic’s “protomap” hypothesis (Rakic, 1988).

A similar understanding of the subpallial-derived interneuron lineages has yet to be achieved. Work from several laboratories, including our own, have demonstrated that inhibitory interneuron populations are entirely derived from the subpallium (reviewed in Fishell and Rudy, 2011), Fig. 1A), most prominently from the medial and caudal ganglionic eminences (MGE and CGE, respectively). In mice, the MGE produces 70 % of all cortical interneurons and the entirety of both the parvalbumin and somatostatin populations, not only within the cortex (Miyoshi et al., 2007) but also in the associated hippocampus (Tricoire et al., 2011), as well as in subpallial structures such as the striatum (Marin et al., 2000) and amygdala nuclei (Nery et al., 2002).

Figure 1. Targeted infection of MGE progenitor cells with a retroviral library.

Figure 1

Open in a new tab

A) Diagram summarizing the experimental design of this study. Orange depicts the area of Nkx2.1-cre expression. B) Experimental strategy. C) Overview images (black and white) and high-magnification images (green) of sections of embryonic mouse brains two to three days after injection of a retroviral library at E10.5 or E12.5. Radial clusters of cells expressing GFP (green, asterisks) were found in the VZ and SVZ of the MGE but not the CGE, LGE or cortex. Several cells with short processes arrayed along a radial process (white arrows). D) Three-dimensional reconstruction illustrating the distribution of GFP-expressing neurons with a barcode in the forebrain of a P16 mouse that was infected with a retroviral library at E10.5. E) Representative high-magnification images and their location within a 25 μm coronal brain section in the cortex (cyan tracing) and hippocampus (yellow tracing). Ctx, cortex; Hi, hippocampus; Scale bars, 500 μm; see also Fig. S1.

Two recent studies have attempted to connect the lineage relationship of MGE-derived interneurons to their cell type and position (Brown et al., 2011; Ciceri et al., 2013). These groups used mouse genetics to specifically label interneuron progenitor cells in the MGE with fluorescently tagged retroviruses. Their results suggested that interneurons arising from a common progenitor preferentially form clusters either within cortical layers and/or columns. An inherent limitation of this approach is that dispersed interneurons labeled with the same fluorophore were assumed ab initio to be derived from independent clones.

To overcome this drawback, we have reinvestigated this question using a barcoded GFP-expressing retroviral library (Golden et al., 1995), restricted in its infection to progenitor cells of the MGE that express the avian virus receptor, TVA, as dictated by a Cre-dependent TVA reporter allele (Seidler et al., 2008).

To our considerable surprise, while a minority of clones were found to be densely clustered, the majority of MGE clones dispersed over large areas both within and across telencephalic structures. Complementing these efforts, we analyzed the acute dispersion of clones using both retroviral and MADM (mosaic analysis with double markers) approaches in the context of short-survival periods. These experiments indicated that most MGE lineages originate from radial glia progenitors in the MGE ventricular zone (VZ) and can be locally amplified in the subventricular zone (SVZ). We conclude that while sibling interneurons may occasionally form coherent clusters, the final positions of clonally related interneurons are generally not constrained by lineal relationships.

Results

Selective targeting of interneuron progenitor cells within the MGE with a barcoded retroviral library

To identify the cortical interneuron populations produced by individual MGE progenitor cells, we used ultrasound-backscatter microscopy (UBM) guided microinjection to inject a retroviral library into the lateral ventricles of transgenic E10.5–E12.5 embryos, the beginning of the peak phase of interneuron neurogenesis in this region (Miyoshi et al., 2007; Fig. 1A, B). The viral library encodes GFP and contains approximately 10ˆ5 random 24 bp oligonucleotide tags (barcodes). To retrieve barcodes, infected cells can be identified by GFP and collected from brain sections by laser capture microdissecion (LCM). Due to the high complexity of the library, and the relatively low number of clones, cells sharing the same barcode are almost certainly siblings (Golden et al., 1995; Fuentealba et al., 2015). The retroviral library used here was pseudotyped with the ASLV-A envelope glycoprotein (EnvA), which restricts infection to cells expressing the cognate avian virus receptor for this glycoprotein, TVA (Bates et al., 1993). We conditionally expressed TVA in mitotic cells of the MGE by crossing an MGE-specific transgenic Cre driver line (Nkx2.1-Cre; Xu et al., 2008) with R26-TVAiLacZ (Seidler et al., 2008) mice. As predicted, two to three days after the injection of the retroviral library isolated radially oriented clusters of GFP-expressing cells were observed in the VZ of the MGE, but not in other progenitor zones of the subpallium or neocortex (Fig. 1C). No GFP labeling was observed in the brains of animals that did not encode TVA (Nkx2.1+/+;R26LSL-TVAiLacZ/+ littermates, not shown). To study the distribution and fate of cells derived from Nkx2.1-Cre expressing cells, Nkx2.1Cre/+;R26TVAiLacZFlox/+ Nkx2.1Cre/+;R26LSL-TVAiLacZ/+ embryos were injected with the retroviral library at E10.5 and sacrificed at P16, a time point by which migration of interneurons is largely complete (Corbin et al., 2001). At P16, GFP-positive neurons were found in a variety of regions throughout the telencephalon (Fig. 1D, 1E, S1), including the neocortex, striatum, olfactory bulb, hippocampus, globus pallidus, hypothalamus and septum, as well as in oligodendrocytes of subcortical structures, consistent with the findings from Nkx2.1 genetic fate-mapping studies (Kessaris et al., 2006; Xu et al., 2008). The barcodes from 47–90 (61 ± 14, N = 3; Fig. 2A) GFP-positive neurons were sequenced. 44 % (± 16 %; N = 3) of the analyzed neurons carried a unique barcode, indicating that they were either single-cell clones or that we failed to sample a sibling cell. 56 % (± 16 %, N = 3; Fig. 2A) of the barcodes, however, were members of multi-cell clones, ranging in size from 2 sibling cells (by far the majority, 52 ± 14 %, N = 3), up to 6 sibling cells (12 ± 12 %, N = 3; Fig. 2C).

Figure 2. Dispersion of interneuron clones across forebrain structures.

Figure 2

Open in a new tab

A) Quantification of single-cell and multi-cell clones identified by barcode analysis. (B) Histogram illustrating the probability in percent that barcoded neurons contribute to a clone of a particular size. (C) Categorization of clones according to their spread. Ctx, cortex; Hi, hippocampus; GP, globus pallidus; N = 3 brains. D) 3D-reconstructions of five clones (green and red dots) illustrating different modes of dispersion.

Dispersion of clonally related interneurons across brain structures

To systematically examine the distribution of clonally related interneurons (i.e., neurons with identical barcodes), the Cartesian coordinates of neurons within the cortex, hippocampus, striatum and globus pallidus from which barcodes were recovered were determined. More than fifty percent of GFP labeled multi-cellular clones were located in the cortex (61 ± 20 % cortex; Fig. S2, 7 ± 6 % hippocampus, 3 ± 6 % striatum; N = 3; Fig. 2C). Strikingly, a number of clonally related neurons were not restricted to only one structure, but rather, were distributed across anatomically and functionally distinct regions of the forebrain. For example, some clones populated the cortex and hippocampus (20 ± 3 %, N = 3), the cortex and striatum (6 ± 6 %, N = 3) or the cortex and globus pallidus (2 % ± 3 %, N = 3; Fig. 2C). By contrast, clones of the telencephalon did not share barcodes with clones in the hypothalamus or contralateral hemisphere, indicating that MGE-derived neurons are restricted to the ipsilateral telencephalon (data not shown). Moreover, identical barcodes were very rarely recovered within different retrovirally infected brains, indicating that the complexity of the library is sufficient to unambiguously resolve lineage relationships. In an attempt to assess the subtype identity of infected neurons we did immunocytochemical analysis in a subset of clones analyzed. In accordance with previous findings (Brown et al., 2011; Ciceri et al., 2013) the existence of clones consisting of entirely parvalbumin positive clones, as well as of mixed parvalbumin positive and negative clones (presumably the majority of the later were somatostatin neurons (Miyoshi et al., 2007)) were observed (Fig. S3).

Dispersion of clonally related interneurons within brain structures

If the lineage relationship contributed to the functional organization of inhibitory interneurons within the mammalian neocortex, one would expect that interneuron clones would reside in close proximity and within similar functional units of the cortex (Yu et al., 2009). To test whether clonally related interneurons settle within focal areas we calculated the Euclidean distances between pairs of sibling cells within these structures (Fig. 3A) and analyzed their distribution (Fig. 3B). Pairs of clonally related neurons were on average almost 2000 μm apart (1947 ± 1274 μm; N = 49 pairs of neurons) and only 8 % of the pairs were located within 500 μm of each other. The average distance between clonally related cells was 1885 μm (± 1132 μm N = 20 clones) in the cortex, 1517 (± 554 μm; N = 5 clones) in the hippocampus and 982 μm (± 495 μm; N = 3 clones; Fig. 3C) in the striatum. This analysis shows that sibling interneurons reside in a volume exceeding functional cortical units, such as whisker barrels of the somatosensory cortex (400 μm; Bruno et al., 2003; Mountcastle, 2003).

Figure 3. Dispersion of clonally related interneurons within brain structures.

Figure 3

Open in a new tab

A) Pairwise distances of all multi-cell clones that were restricted to one brain structure are shown for three experiments (three analyzed brains). B) Histogram of pairwise distances for pooled data of three experiments. Bin-size, 500μm. C) Average distances between neurons within clones in the cortex, hippocampus and striatum. D) Dendrograms showing hierarchical relationships between clones of one brain structure grouped according to their Euclidian distances and labeled according to their lineal relationship. Numbers and colors indicate the lineal relationship between two or more cells based on barcode-sequencing. Colored lines below numbers mark isolated clonal clusters; colored dots mark split sibling neurons of the latter. E) Quantification of the distance between isolated clonal clusters. F) Nearest neighbor distance and average distance of multi-cell clones of one structure (clones) and all labeled neurons with a barcode (non-clones). See also Fig. S2, S4.

Next we asked whether lineal relationship predicts the position of sibling cells within the cortex, hippocampus or striatum. We reasoned that if this was the case, then clonally related neurons should preferentially be clustered compared to members of unrelated interneuron lineages. To test this, we examined the lineage relationship of localized clusters. We first grouped GFP labeled neurons irrespective of their lineage by their proximity and displayed the results in dendrograms (Fig. 3D, S4). We then labeled the neurons according to their lineage relationship (i.e., barcode identity). 15 out of 27 clones did not form isolated clusters. The isolated clonal clusters that we did observe (Fig. 3D, 3E; #7, #8, #11, #16, #21, #22, #24, #25, #26, #28, #30) were mainly 2-cell clusters, of which four (clone #11, #25, #26 and #31) had an additional sister cell outside the isolated cluster (splitting). The average distance between isolated clonally related cells (splitting cells not included) was 818 μm (± 589 μm, N = 12 cluster, Fig. 3E). Next we included single-cell barcodes (i.e., neurons with unique barcodes) into the analysis to increase the pool of interneurons and calculated the nearest neighbor distance (NND) and the average distance (AD) between pairs of neurons. As expected, the NND of unrelated neurons was significantly smaller (560 ± 337 μm; N = 185 interneuron pairs) than of sibling neurons (1550 ± 965 μm; N = 84; P < 0.0001; Fig. 2E). Strikingly, there was no significant difference in the AD between pairs of sibling neurons and pairs of unrelated neurons (1498 ± 617.5 μm; N = 17; vs. 1723 ± 1027 μm; N = 28; P = 0.8; Fig. 3F). Furthermore, no significant difference was detected in the spatial separation of the NND vs. the AD between clones (Fig. 3F), indicating that splitting of clonal clusters into two or more clusters is not a common mode of organization. Grouping members of clones according to their proximity into dendrograms resulted in only five isolated clusters when single cells were included in the analysis (Fig. S4).

Organization of interneuron progenitor cells in the MGE

To interpret the significance of our findings, it is essential to determine the mode of cell division that progenitors undergo after retroviral labeling. We examined the lineages of newly postmitotic interneurons within selected samples of GFP-positive cells in the VZ and SVZ, two to three days post-infection with the retroviral library. Because individual cells within dense clusters of GFP positive cells could not be separated using LCM, cells were captured as groups. PCR-fragments harboring the barcodes were subcloned into plasmids and transformed into competent bacterial cells. Sequencing the barcodes of multiple bacterial colonies allowed an estimation of the number of barcodes per cluster, and the identification of their sequence. Fig. 4B shows representative images illustrating the clonal organization within the MGE. Similar to published observations (Brown et al., 2011), we found evidence for radially aligned clones, with one cell touching the ventricular surface and additional cells being symmetrically aligned in close vicinity (Fig. 4B1, B2; N = 6), likely demonstrating asymmetrical progenitor divisions (Fig. 4A; Brown et al., 2011; Noctor et al., 2001). We also found monoclonal clusters in the SVZ that were often attached to radial glia fibers (Fig. 4B3, B4; N = 9), suggesting that symmetrical terminal divisions occur in the SVZ (Brown et al., 2011; Noctor et al., 2004). With increasing distance from the ventricular surface, clonal boundaries became indistinct and were impossible to predict. Patches of laser-captured tissue from such samples contained multiple cells with multiple barcodes (Fig. 4B5; N = 9). At this stage, cells were not attached to radial glia fibers, likely indicating the mixing of distinct lineages as a result of the migration of neuronal precursor cells. We did not find evidence for shared barcodes between radial glia, nor did we observe symmetric radial glial divisions. Taken together our results suggest that clonal dispersion throughout the forebrain arises from asymmetrical progenitor divisions, accompanied by symmetrical terminal divisions within the SVZ.

Figure 4. Spatially organized clonal units in the MGE.

Figure 4

Open in a new tab

A) Diagram illustrating different kinds of cell divisions that have previously been described in the MGE of embryonic mice (Brown et al., 2011). B) High magnification images of groups of neurons (green) and their exact location depicted within low magnification overview images of coronal brain sections (black and white images). Rectangles indicate the location of high-magnification images; yellow circles and labels indicate the clonal relation between groups of neurons (i.e., monoclonal clusters of polyclonal cell groups). C) Representative examples of E10–12 and E10–13 MADM clones in the VZ/SVZ of the MGE. The total clone size was 11,1 ± 1,389 (N = 9) for E10–E12, and 21 ± 2,188 (N = 8) for E10–13, respectively. The majority of G2-X MADM clones (15/17) display unequal lineage trees with respect to the absolute number of labeled neurons in the two (red and green) sub-lineages. Scale bars, 500 μm (B); 50 μm (C); see also Fig. S5.

To independently assess the mode of divisions within MGE progenitors at this developmental stage, we analyzed clones at different time points using the MADM method (Zong et al., 2005), which provides single cell resolution of progenitor division pattern. A key MADM feature is the ability to induce clones of distinctly labeled neurons originating from a single dividing progenitor cell in a temporally defined fashion using tamoxifen (TM) inducible CreER driver lines (Bonaguidi et al., 2011; Hippenmeyer et al., 2010; Zong et al., 2005; Fig. S5). To achieve selective labeling of interneuron progenitors and their progeny, we used MADM-11 in combination with Nestin-CreERT2 (Hippenmeyer et al., 2010). A single TM dose was administered to timed pregnant Nestin-CreERT2/MADM-11 dams at E10 via intraperitoneal injection. Embryonic brains were recovered two or three days post TM injection, processed for serial cryosectioning and immunostained to visualize all MADM-labeled cells in the VZ/SVZ of the MGE. The MADM-labeling efficiency of MGE progenitors was very low (less than one clone per brain on average) and in the absence of TM treatment, we found no labeled cells. MADM clones in the MGE displayed radially arrayed clusters of cells in the VZ (Fig. 4C) and clumpy clusters of cells in the SVZ, both found in close proximity to a radial glia process. Consistent with our retroviral findings, individual clusters contained either a combination of green and red fluorescent neurons (G-X clone, Fig. 4C) or yellow fluorescent neurons only (G2-Z or G1 events; data not shown). The majority of MADM clones (15/17) displayed labeling consistent with asymmetric neurogenic divisions, as indicated by the presence of an unequal amount of red and green cells within a cluster. In addition, a much smaller number (2/17) of MGE progenitors produced symmetric lineage trees within the 2–3 day developmental time window (Fig. 4C, upper right panel). Within the distinctly labeled MADM subclones we frequently observed clusters of labeled cells in the SVZ similar to the above clones labeled with retrovirus. Upon exiting the MGE SVZ, MADM labeled cells dispersed widely and migrated towards the dorsal telecephalon. This corroborates our interpretation that MGE progenitors undergoing active neurogenesis produce widely dispersed progeny that contribute to a variety of telencephalic structures in a seemingly unconstrained manner. Our data suggest that: 1) retrovirally labeled clones of future neocortical and hippocampal interneurons in the MGE are initially organized into radial arrays, similar to clones of excitatory neurons (Brown et al., 2011; Rakic, 1988); and 2) intermediate progenitors in the MGE SVZ further amplify the number of postmitotic interneurons through symmetric neurogenic divisions (Fig. 4A; Noctor et al., 2004).

Dispersion of clonally related cells after uncovering transcriptionally silenced viral vectors

The number of cells per clone was surprisingly large in short term analysis (Fig. 4C) compared with clones examined after migration (Fig. 2B). Previous lineage studies (Cepko et al., 2000; Halliday and Cepko, 1992; McCarthy et al., 2001) in the forebrain have noted transcriptional silencing of retroviral vectors. As we identified barcoded neurons based on their expression of GFP, transcriptional silencing of GFP by P16 could explain the observed discrepancy in clone size. To recover putative barcodes regardless of silencing, we collected large pieces of tissue from brain sections of P16 mice that were infected with the retroviral library at E10.5. From the same sections in which all GFP-expressing neurons had previously been individually collected via LCM (Exp. 3 in Fig. 2A and Fig. 3A), we processed 228 pieces of cortical, hippocampal and striatal tissue. These were collected from individual 25 μm coronal brain sections to maintain the spatial resolution of neurons with silenced barcodes. Barcodes were PCR amplified, ligated into plasmids and transformed into bacterial cells. Sequencing of a large number of plasmids (1026, of which the vast majority was repetition of identical barcodes due to over-sampling of bacterial colonies), each isolated from a single bacterial colony, indicated that every coronal brain section contained on average 1.8 ± 1.4 silenced barcodes per structure (n = 3 structures). The results were added to the existing dataset of identified barcodes, increasing the number of recovered barcodes from 47 to 406 (Fig. 5C, Fig. 2A). These data indicate that a large amount of retroviral silencing has occurred by P16. Recovering silenced barcodes reduced the percentage of single-cell clones from 44 ± 16 % (N = 3 brains) to 32 % (N = 1 hemisphere; Fig. 5B), suggesting that many singletons in the previous experiments had siblings with silenced vectors. When clones were plotted according to their location along the anterior-posterior axis (Fig. 5C), a large spread of clonally related interneurons was evident. Despite adding a few large clones (including one 13-cell and one 15-cell clone) the number of neurons per clone did not markedly change (Fig. 5D, Fig. 2B). In addition, the relative distribution of clones within and across different structures of the forebrain was similar when silenced clones were added to the analysis (Fig. 5E). Notably, the AD between pairs of related cells was similar to unrelated cells. Hence, these results provide further support for the widespread dispersion of MGE-derived clones within the ipsilateral telencephalon.

Figure 5. Distribution of transcriptionally silenced viral vectors.

Figure 5

Open in a new tab

A) Schematic illustrating the strategy to recover silenced barcodes. Note: the spatial resolution of silenced barcodes is maintained only by anatomical structure and within the anterior-posterior axis. B) Quantification of single-cell and multi-cell clones identified by barcode analysis. C) The location of both GFP-positive and GFP-negative infected neurons is plotted relative to Bregma on the x-axis. Dots connected with a horizontal line represent the location of sibling neurons. Left, sibling cells that were dispersed either within the cortex (red), hippocampus (blue) or striatum (green); right, barcodes that represent sibling cells that were spread across structural boundaries. Relative clonal size (D) and spread both within and across brain structures (E). F) AD of multi-cell clones and single-cell clones in different brain structures.

Discussion

Our analysis provides a description of the relationship between MGE-derived interneuron lineages and their global distribution within the telencephalon. Examination of the final position of MGE-derived interneuron clones within the brain revealed a dramatic dispersion of sister cells both within and across structural boundaries within the telencephalon. By contrast, clones were not seen to cross the segmental boundary of the diencephalon and telencephalon, or the midline between the two cerebral hemispheres. While it remains possible that some clones occupy small functional units in the forebrain, they would be in the minority, as most clones observed here were widely distributed.

A fundamental parameter needed to interpret these results was the mode of division that the infected progenitor cells underwent subsequent to retroviral integration. If the clones resulted from symmetric self-renewing stem-cell (i.e., non-neurogenic) divisions, it would perhaps not be surprising that a high degree of dispersion was observed. By contrast, if the clones were produced from asymmetric neurogenic divisions, it would imply an unexpected ability for lineage related clones to be allocated to distinct structures. We utilized two independent approaches to ascertain the mode of cell division. First, we examined the distribution of infected cells containing retroviral barcodes one to three days post-infection, second, we utilized MADM (Zong et al., 2005), in which recombination in progenitor cells leads to the production of green or red fluorescent protein in the two daughter cells. The results from both analyses supported a model in which the majority of MGE-derived interneurons were generated by neurogenic amplification divisions.

In comparing the short-term and long-term analysis of clones, we observed that the clone size differed substantially. We speculate that the relatively small clone size observed at P16 is attributable to a combination of significant cell death occurring during interneuron maturation (Southwell et al., 2012), a degree of failure in the capture of sibling cells, and as a result of progressive increases in retroviral silencing with age. In contrast to studies relying purely on the expression of retroviral marker genes, the genomic barcodes allowed us to estimate the amount of silencing and even to include barcodes obtained from silenced vectors into our analysis (although doing so came at the price of reducing the spatial resolution compared to neurons captured based on GFP-expression, because the position of silenced clones is only as precise as the size of the tissue excised by LCM). While our results clearly indicate a large amount of silencing at P16, they also show that silencing does not warp the results from GFP labeled neurons with barcodes. Silenced clones showed a similar spread within and across brain structures as GFP-labeled clones and at least in aggregate consistently did not reside as a single cluster. Notably, the average clone size was (despite few very large clones) very similar with and without silenced clones, indicating that clones are preferentially silenced on an ‘all or nothing’ basis, likely based on the position within the genome of retroviral insertion. Nonetheless, the quantification of silencing points out that our analysis at least with regard to the recovery of large clones systematically underestimates the number of lineally related neurons. As such, it will be important in the future to develop new retroviral vectors that are less prone to silencing.

Our findings seem to contradict two recent studies (Brown et al., 2011; Ciceri et al., 2013), that suggested that MGE-derived clones form spatially isolated clusters in cortical columns or laminae. Both studies used low-titer retrovirus infections and defined clonality of post-migratory neurons based on geometric criteria. This unavoidably results in both lumping errors (clustered cells that are not clonal) and splitting errors (dispersed cells that are clonal but not recognized as such). While our results rule out that all members of a clone are preferentially clustered, we observed that subsets of clones in some instances reside in close proximity and form isolated clusters, consistent with the previous studies. As such, although clearly not an absolute rule, it remains possible that the members of a subset of clones have a spatial and perhaps functional relationship in the mature brain. Conversely, given that clonally related cells are produced in an environment that might expose them to similar guidance or other environmental cues, they may become clustered in a final location not due to lineage, but as a result of common environmental cues that guide their migration.

Work from a variety of organisms supports that lineage contributes to the generation of cell diversity. The role of lineage in invertebrate (Hobert, 2010) and vertebrate species has been both studied and compared (Cepko, 2014). In Drosophila and nematodes it is very clear that specific lineages can generate predictable but highly divergent cell types (Hobert, 2010). In vertebrates, fewer studies have been carried out and lineage descriptions are much less comprehensive. However, clonal analysis in the retina has suggested that at least near terminal lineages may also be stereotyped (Godinho et al., 2007). However, whether these are derived from larger restricted lineages is not clear (Cepko, 2014). Does the immense diversity of regionally specified cell populations within different brain circuits result from fate-restricted lineages? Given the vast expansion in neuronal numbers in the brain of mammals (Geschwind and Rakic, 2013; Molnár and Clowry, 2012), this is an attractive hypothesis. Our own efforts and those of others to examine the origins of interneurons within the forebrain demonstrate that specific progenitor zones, the MGE in particular (reviewed in Fishell and Rudy, 2011), give rise to the large majority of cortical and hippocampal interneuron populations. Similar work strongly suggests that subpallial structures such as the striatum and amygdala derive their interneuron populations from the same embryonic sources (Marin et al., 2000; Nery et al., 2002). While certain commonalities exist, there are also marked differences in the abundance, intrinsic properties and connectivity of interneurons within different telencephalic regions (Kawaguchi et al., 1995; Klausberger and Somogyi, 2008; Kubota and Kawaguchi, 1994).

These observations have prompted the question as to how such specificity is achieved and whether lineage restrictions play a role in this process. Several studies have demonstrated that the specification of interneuron subtypes is initiated during proliferative phases within the progenitor zones (Butt et al., 2008). However, region-specific migration of interneurons can still be altered during postmitotic stages (e.g. McKinsey et al., 2013; van den Berghe et al., 2013). Thus, the expression of genes affecting the positioning of cells can potentially be dictated by lineage, as well as induced postmitotically by environmental cues. For instance, recent work indicates that electrical activity influences the migration of postmitotic interneurons (Bortone and Polleux, 2009; De Marco García et al., 2011), their morphological development and their connectivity (Spiegel et al., 2014). Our results are consistent with environmental cues and stochastic choices affecting interneuron positioning independent of lineage.

Although not restricting sibling cells to specific structures, lineage may still contribute to the generation of interneuron diversity. For instance, one could imagine that specific lineages could create progeny that share a common program that is contextually modified after migration is completed. In such a model, the dispersion of a common pool of progenitors across structures could allow for different regions of the telencephalon to acquire interneurons with specific properties, while permitting them to adjust their functional program in accordance with the requirements of particular brain structures (Kepecs and Fishell, 2014). Furthermore, the MGE does not solely produce interneuron populations (Flandin et al., 2010; Nery et al., 2002). In vivo fate mapping of the MGE has demonstrated that progenitor cells within this region give rise to GABAergic projection neurons of the globus pallidus (Flandin et al., 2010), as well as portions of both the nucleus accumbens and amygdala (Nery et al., 2002). An important goal of future analysis will be to explore if there is a predictable lineage relationship between the diverse types of MGE-derived GABAergic populations.

In summary, our work demonstrates that individual MGE-derived lineages contribute to broad areas and distinct structures within the telencephalon. This indicates that regionally specific interneurons found in different brain circuits are not generated by dedicated progenitors within the MGE. However, whether an important role for lineage exists in the creation of specific interneuron populations and the underlying logic by which such lineages create diversity remains a possibility that is well worth exploring.

Experimental Procedures

Sample Collection (Laser Capture)

GFP-positive cells were collected individually using a laser microdissection system (LMD6000, Leica). Cells were collected into 20 μl of lysis buffer including Proteinase K (1:1000; Qiagen). Small patches of GFP-negative tissue next to the collected cell were frequently collected as a control.

Barcode Sequencing

Cells were lysed and viral barcodes were PCR amplified from the viral vector via a two step nested PCR. PCR conditions for the first PCR were: 60°C annealing temperature with 35 seconds elongation time using 40 cycles. Primer sequences were SBR161-o1, gacaaccactacctgagcacccagt and SBR126-o2, ggctcgtactctataggcttcagctggtga. PCR conditions for the nested PCR were: 60°C annealing temperature with 35 seconds elongation time using 30 cycles. Primer sequences were SBR160-n1, atcacatggtcctgctggagttcgtga and SBR128-n2, attgttgagtcaaaactagagcctggacca. PCR products were visualized, gel-purified (Gel Band Purification Kit, GE Healthcare) and sequenced.

Barcode recovery from silenced vectors

To recover barcodes from retrovirally infected cells that did not express GFP due to silencing, we collected large pieces of tissue from the same PET membrane slides using LCM. Tissues collected from different sections and different brain structures were processed separately to maintain spatial information of barcodes. Individual barcodes were sequenced from a large number of plasmids, each isolated from a single bacterial colony. To estimate the number of barcodes per tissue, a minimum of 6 colonies were sequenced per sample. If more then one barcode was present then up to 16 colonies were sequenced per sample. In a small number of instances (8 of 1210 barcodes recovered in total), the same barcode was recovered in more than one brain and hemisphere. We assumed that this was a result of contamination or overrepresentation of that particular barcode in the retroviral library and therefore any lineages marked by these 8 barcodes were excluded from the analysis.

MADM Clone Induction

Embryonic MADM interval clones were generated as described previously (Hippenmeyer et al., 2010). Pregnant MADM-11GT/TG;Nestin-CreER+/− females were injected intraperitoneally with a maximal dose of 2–3mg tamoxifen (TM, dissolved in corn oil, Sigma) at E10. Embryos (MADM-11GT/TG;Nestin-CreER+/−) were isolated at E12 or E13, respectively, and brains were fixed in 4 % PFA/PBS for 2–4h following cryopreservation in 30 % sucrose/PBS. Cryosections 30 μm thick were obtained using a cryostat (Microm). The GFP and tdT signal was amplified by antibody staining and nuclei visualized using DAPI (Invitrogen). MADM clones in the MGE were imaged with a Zeiss LSM700 confocal microscope and the total MADM clone size (#red + #green cells; average ± SEM) was determined for E10-E12 (n=9) and E10–13 (n=8) clones, respectively. The efficiency of MADM clone induction was much lower in the ventral than in the dorsal telecephalon and less than one MGE clone per brain was observed on average. In the absence of TM, no MADM labeling was observed.

Data Analysis

The average distance (i.e., the average distance between every pair of sibling neurons) and nearest neighbor distance between sibling neurons was calculated after 3D-reconstruction of the brain in Neurolucida software (MBF Bioscience). Cartesian coordinates of every barcode within the forebrain were exported from Neurolucida to Matlab software (Mathworks), to calculate Euclidian distances between pairs of sibling neurons. A hierarchical, binary cluster tree was created by the linkage function and plotted into dendrograms. Data are presented as mean ± s.e.m. and nonparametric tests (Mann-Whitney-Wilcoxon) were used for statistical significance estimations in Prism software (Graphpad).

Supplementary Material

1
2

Acknowledgments

We are grateful to Robert Machold, Niels Ringstad and Richard Tsien for comments on the manuscript. Research in the Fishell laboratory is supported by NIH (NS 081297, MH095147, P01NS074972), the Simons Foundation. Research in the Hippenmeyer laboratory is supported by the European Union (FP7-CIG618444 to S.H.). C.M. is supported by EMBO ALTF 1295–2012. X.J. is supported by EMBO (ALTF 303–2010) and HFSP (LT000078/2011-L).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Bates P, Young JAT, Varmus HE. A receptor for subgroup A Rous sarcoma virus is related to the low density lipoprotein receptor. Cell. 1993;74:1043–1051. doi: 10.1016/0092-8674(93)90726-7. [DOI] [PubMed] [Google Scholar]
  2. Bonaguidi MA, Wheeler MA, Shapiro JS, Stadel RP, Sun GJ, Ming GL, Song H. In vivo clonal analysis reveals self-renewing and multipotent adult neural stem cell characteristics. Cell. 2011;145:1142–1155. doi: 10.1016/j.cell.2011.05.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bortone D, Polleux F. KCC2 expression promotes the termination of cortical interneuron migration in a voltage-sensitive calcium-dependent manner. Neuron. 2009;62:53–71. doi: 10.1016/j.neuron.2009.01.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brown KN, Chen S, Han Z, Lu CH, Tan X, Zhang XJ, Ding L, Lopez-Cruz A, Saur D, Anderson SA, et al. Clonal production and organization of inhibitory interneurons in the neocortex. Science. 2011;334:480–486. doi: 10.1126/science.1208884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bruno RM, Khatri V, Land PW, Simons DJ. Thalamocortical angular tuning domains within individual barrels of rat somatosensory cortex. J Neurosci. 2003;23:9565–9574. doi: 10.1523/JNEUROSCI.23-29-09565.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Butt SJB, Sousa VH, Fuccillo MV, Hjerling-Leffler J, Miyoshi G, Kimura S, Fishell G. The requirement of Nkx2-1 in the temporal specification of cortical interneuron subtypes. Neuron. 2008;59:722–732. doi: 10.1016/j.neuron.2008.07.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cepko C. Intrinsically different retinal progenitor cells produce specific types of progeny. Nat Rev Neurosci. 2014;15:615–627. doi: 10.1038/nrn3767. [DOI] [PubMed] [Google Scholar]
  8. Ciceri G, Dehorter N, Sols I, Huang ZJ, Maravall M, Marin O. Lineage-specific laminar organization of cortical GABAergic interneurons. Nat Neurosci. 2013;16:1199–1210. doi: 10.1038/nn.3485. [DOI] [PubMed] [Google Scholar]
  9. Corbin JG, Nery S, Fishell G. Telencephalic cells take a tangent: non-radial migration in the mammalian forebrain. Nat Neurosci. 2001;4:1177–1182. doi: 10.1038/nn749. [DOI] [PubMed] [Google Scholar]
  10. De Marco García NV, Karayannis T, Fishell G. Neuronal activity is required for the development of specific cortical interneuron subtypes. Nature. 2011;472:351–355. doi: 10.1038/nature09865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fishell G, Rudy B. Mechanisms of inhibition within the telencephalon: “where the wild things are”. Annu Rev Neurosci. 2011;34:535–567. doi: 10.1146/annurev-neuro-061010-113717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fuentealba LC, Rompani SB, Parraguez JI, Obernier K, Romero R, Cepko CL, Alvarez-Buylla A. Embryonic Origin of Postnatal Neural Stem Cells. Cell. 2015;161:1644–1655. doi: 10.1016/j.cell.2015.05.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Flandin P, Kimura S, Rubenstein JLR. The progenitor zone of the ventral medial ganglionic eminence requires Nkx2-1 to generate most of the globus pallidus but few neocortical interneurons. J Neurosci. 2010;30:2812–2823. doi: 10.1523/JNEUROSCI.4228-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. García-Moreno F, Vasistha NA, Begbie J, Molnár Z. CLoNe is a new method to target single progenitors and study their progeny in mouse and chick. Development. 2014;141:1589–1598. doi: 10.1242/dev.105254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Geschwind DH, Rakic P. Cortical evolution: judge the brain by its cover. Neuron. 2013;80:633–647. doi: 10.1016/j.neuron.2013.10.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Godinho L, Williams PR, Claassen Y, Provost E, Leach SD, Kamermans M, Wong ROL. Nonapical symmetric divisions underlie horizontal cell layer formation in the developing retina in vivo. Neuron. 2007;56:597–603. doi: 10.1016/j.neuron.2007.09.036. [DOI] [PubMed] [Google Scholar]
  17. Golden JA, Fields-Berry SC, Cepko CL. Construction and characterization of a highly complex retroviral library for lineage analysis. Proc Natl Acad Sci USA. 1995;92:5704–5708. doi: 10.1073/pnas.92.12.5704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hippenmeyer S, Youn YH, Moon HM, Miyamichi K, Zong H, Wynshaw-Boris A, Luo L. Genetic mosaic dissection of Lis1 and Ndel1 in neuronal migration. Neuron. 2010;68:695–709. doi: 10.1016/j.neuron.2010.09.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hobert O. Neurogenesis in the nematode Caenorhabditis elegans. WormBook. 2010:1–24. doi: 10.1895/wormbook.1.12.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kawaguchi Y, Wilson CJ, Augood SJ, Emson PC. Striatal interneurones: chemical, physiological and morphological characterization. Trends Neurosci. 1995;18:527–535. doi: 10.1016/0166-2236(95)98374-8. [DOI] [PubMed] [Google Scholar]
  21. Kepecs A, Fishell G. Interneuron cell types are fit to function. Nature. 2014;505:318–326. doi: 10.1038/nature12983. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kessaris N, Fogarty M, Iannarelli P, Grist M, Wegner M, Richardson WD. Competing waves of oligodendrocytes in the forebrain and postnatal elimination of an embryonic lineage. Nat Neurosci. 2006;9:173–179. doi: 10.1038/nn1620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Klausberger T, Somogyi P. Neuronal diversity and temporal dynamics: the unity of hippocampal circuit operations. Science. 2008;321:53–57. doi: 10.1126/science.1149381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kohwi M, Doe CQ. Temporal fate specification and neural progenitor competence during development. Nat Rev Neurosci. 2013;14:823–838. doi: 10.1038/nrn3618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kubota Y, Kawaguchi Y. Three classes of GABAergic interneurons in neocortex and neostriatum. Jpn J Physiol. 1994;44:S145–S148. [PubMed] [Google Scholar]
  26. Marin O, Anderson SA, Rubenstein JL. Origin and molecular specification of striatal interneurons. J Neurosci. 2000;20:6063–6076. doi: 10.1523/JNEUROSCI.20-16-06063.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. McKinsey GL, Lindtner S, Trzcinski B, Visel A, Pennacchio LA, Huylebroeck D, Higashi Y, Rubenstein JLR. Dlx1&2-dependent expression of Zfhx1b (Sip1, Zeb2) regulates the fate switch between cortical and striatal interneurons. Neuron. 2013;77:83–98. doi: 10.1016/j.neuron.2012.11.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Miyoshi G, Butt SJB, Takebayashi H, Fishell G. Physiologically distinct temporal cohorts of cortical interneurons arise from telencephalic Olig2-expressing precursors. J Neurosci. 2007;27:7786–7798. doi: 10.1523/JNEUROSCI.1807-07.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Molnár Z, Clowry G. Cerebral cortical development in rodents and primates. Prog Brain Res. 2012;195:45–70. doi: 10.1016/B978-0-444-53860-4.00003-9. [DOI] [PubMed] [Google Scholar]
  30. Mountcastle VB. Introduction. Computation in cortical columns. Cereb Cortex. 2003;13:2–4. doi: 10.1093/cercor/13.1.2. [DOI] [PubMed] [Google Scholar]
  31. Nery S, Fishell G, Corbin JG. The caudal ganglionic eminence is a source of distinct cortical and subcortical cell populations. Nat Neurosci. 2002;5:1279–1287. doi: 10.1038/nn971. [DOI] [PubMed] [Google Scholar]
  32. Noctor SC, Flint AC, Weissman TA, Dammerman RS, Kriegstein AR. Neurons derived from radial glial cells establish radial units in neocortex. Nature. 2001;409:714–720. doi: 10.1038/35055553. [DOI] [PubMed] [Google Scholar]
  33. Noctor SC, Martínez-Cerdeño V, Ivic L, Kriegstein AR. Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases. Nat Neurosci. 2004;7:136–144. doi: 10.1038/nn1172. [DOI] [PubMed] [Google Scholar]
  34. Rakic P. Specification of cerebral cortical areas. Science. 1988;241:170–176. doi: 10.1126/science.3291116. [DOI] [PubMed] [Google Scholar]
  35. Seidler B, Schmidt A, Mayr U, Nakhai H, Schmid RM, Schneider G, Saur D. A Cre-loxP-based mouse model for conditional somatic gene expression and knockdown in vivo by using avian retroviral vectors. Proc Natl Acad Sci USA. 2008;105:10137–10142. doi: 10.1073/pnas.0800487105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Southwell DG, Paredes MF, Galvao RP, Jones DL, Froemke RC, Sebe JY, Alfaro-Cervello C, Tang Y, Garcia-Verdugo JM, Rubenstein JL, et al. Intrinsically determined cell death of developing cortical interneurons. Nature. 2012;491:109–113. doi: 10.1038/nature11523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Spiegel I, Mardinly AR, Gabel HW, Bazinet JE, Couch CH, Tzeng CP, Harmin DA, Greenberg ME. Npas4 regulates excitatory-inhibitory balance within neural circuits through cell-type-specific gene programs. Cell. 2014;157:1216–1229. doi: 10.1016/j.cell.2014.03.058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tricoire L, Pelkey KA, Erkkila BE, Jeffries BW, Yuan X, McBain CJ. A blueprint for the spatiotemporal origins of mouse hippocampal interneuron diversity. J Neurosci. 2011;31:10948–10970. doi: 10.1523/JNEUROSCI.0323-11.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. van den Berghe V, Stappers E, Vandesande B, Dimidschstein J, Kroes R, Francis A, Conidi A, Lesage F, Dries R, Cazzola S, et al. Directed migration of cortical interneurons depends on the cell-autonomous action of Sip1. Neuron. 2013;77:70–82. doi: 10.1016/j.neuron.2012.11.009. [DOI] [PubMed] [Google Scholar]
  40. Vasistha NA, García-Moreno F, Arora S, Cheung AFP, Arnold SJ, Robertson EJ, Molnár Z. Cortical and Clonal Contribution of Tbr2 Expressing Progenitors in the Developing Mouse Brain. Cereb Cortex bhu. 2014:125. doi: 10.1093/cercor/bhu125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Walsh C, Cepko CL. Widespread dispersion of neuronal clones across functional regions of the cerebral cortex. Science. 1992;255:434–440. doi: 10.1126/science.1734520. [DOI] [PubMed] [Google Scholar]
  42. Xu Q, Tam M, Anderson SA. Fate mapping Nkx2.1-lineage cells in the mouse telencephalon. J Comp Neurol. 2008;506:16–29. doi: 10.1002/cne.21529. [DOI] [PubMed] [Google Scholar]
  43. Yu YC, Bultje RS, Wang X, Shi SH. Specific synapses develop preferentially among sister excitatory neurons in the neocortex. Nature. 2009;458:501–504. doi: 10.1038/nature07722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zong H, Espinosa JS, Su HH, Muzumdar MD, Luo L. Mosaic analysis with double markers in mice. Cell. 2005;121:479–492. doi: 10.1016/j.cell.2005.02.012. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS717516-supplement-1.docx (566.9KB, docx)
2
NIHMS717516-supplement-2.pdf (44.5MB, pdf)

RESOURCES