TY - JOUR
AU - Luebke, Jennifer, I
AB - Abstract How the variety of neurons that organize into neocortical layers and functional areas arises is a central question in the study of cortical development. While both intrinsic and extrinsic cues are known to influence this process, whether distinct neuronal progenitor groups contribute to neuron diversity and allocation is poorly understood. Using in vivo genetic fate-mapping combined with whole-cell patch clamp recording, we show that the firing pattern and apical dendritic morphology of excitatory neurons in layer 4 of the barrel cortex are specified in part by their neural precursor lineage. Further, we show that separate precursors contribute to unique features of barrel cortex topography including the intralaminar position and thalamic innervation of the neurons they generate. Importantly, many of these lineage-specified characteristics are different from those previously measured for pyramidal neurons in layers 2–3 of the frontal cortex. Collectively, our data elucidate a dynamic temporal program in neuronal precursors that fine-tunes the properties of their progeny according to the lamina of destination. barrel cortex, intermediate progenitor, layer 4, neuron specification, neuronal diversity Introduction The allocation of multiple neuron types across the 6 layers of the neocortex is key for the specification of functionally distinct cortical areas and circuits (Rakic 1988). The diversity of neurons within each layer also contributes to the intricate microcircuitry of the cortex. While the developmental programs underlying these features rely on a dynamic interplay between intrinsic and extrinsic factors (Sur and Rubenstein 2005; Dehay and Kennedy 2007), how neuronal precursor types contribute to this diversity is unknown. In the developing dorsal mammalian telencephalon, at least 4 groups of proliferative cells generate excitatory neurons. Apical radial glial cells (aRGCs), the stem cells of the neocortex, undergo asymmetric cell divisions at the ventricular surface to either directly generate neurons or to produce at least 3 classes of intermediate progenitor cells (IPCs) which contemporaneously produce neurons, a process defined as indirect neurogenesis (Malatesta et al. 2000; Noctor et al. 2001, 2002). Namely, apical intermediate precursor cells (aIPCs, also known as short neural precursors, SNPs) reside in the ventricular zone (VZ) and undergo symmetric divisions at the ventricular surface to generate neurons, while basal IPCs (bIPCs) form a secondary proliferative compartment away from the ventricle, the subventricular zone (SVZ), where they too divide symmetrically to generate neurons (Englund et al. 2005; Gal et al. 2006; Mizutani et al. 2007; Kowalczyk et al. 2009; Stancik et al. 2010; Tyler and Haydar 2013; Elsen et al. 2013; Nelson et al. 2013). Basal radial glial cells (bRGCs) also divide away from the ventricle in the SVZ but can be distinguished from bIPCs by virtue of their long basal fiber and ability to undergo limited rounds of self-renewing cell divisions (Fietz et al. 2010; Hansen et al. 2010; Shitamukai et al. 2011; Betizeau et al. 2013). Despite an ever-increasing understanding of the transcriptional basis for this heterogeneity in the neurogenic niche (Kawaguchi et al. 2008; Pollen et al. 2014; Johnson et al. 2015), a crucial question remains: do the diverse populations of intermediate precursor cells merely amplify the neuronal output of RGCs and expand cortical surface area, or do they generate neurons with unique intrinsic properties? In a previous study, we presented evidence for the latter possibility, illustrating that pyramidal neurons within layers 2–3 of mouse frontal cortex exhibit distinct morphological and electrophysiological properties based on lineage (Tyler et al. 2015). A goal of the present study was to examine a different area and layer of the neocortex to determine whether lineage-specified intralaminar diversity is a common principle of neuronal development. To do so, we fate-mapped the progeny of 2 major groups of progenitor cells in the future somatosensory cortex, those that express the transcription factor Tbr2 and those that do not (Tyler and Haydar 2013; Tyler et al. 2015), at a time when layer 4 neurons are generated. The mouse somatosensory cortex provides a unique system for this study for several reasons. First, its prominent and well-characterized laminar and columnar organization (Woolsey and Van der Loos 1970) is ideal for investigation of the developmental processes leading to cortical topography. Layer 4 contains multiple classes of excitatory neurons (Feldmeyer et al. 1999; Schubert et al. 2003; Staiger et al. 2004; Oberlaender et al. 2012), making it an excellent area to study the origins of intralaminar diversity. Finally, layer 4 is particularly well suited to examine the interactions between intrinsic and extrinsic factors that shape the formation of cortical areas, as it is the main recipient of thalamic input, a key determinant of areal specification (O’Leary et al. 1994; Hevner et al. 2002). Our data show that direct and indirect routes of neurogenesis contribute specific features of the columnar topography in the barrel cortex. Furthermore, daughter neurons of separate precursor lineages inherit distinct morphological and electrophysiological properties. Finally, we show for the first time that this lineage-specified identity influences the distribution of thalamic afferent innervation on the dendritic arbors of individual neurons. Our results demonstrate that functional classes of neurons are contemporaneously generated by separate groups of precursors and that each lineage changes temporally to modify the properties of their daughter neurons according to the lamina of destination. Materials and Methods In Utero Electroporation In utero electroporation was performed as described elsewhere (Gal et al. 2006) at gestational age E13.5 on timed pregnant CD1 mice from Charles River Laboratories. The Tbr2-Cre and CAG-Stoplight plasmids, or Tbr2-Cre and Tbr2-Stoplight plasmids, were mixed at a 1:1 ratio by copy number and mixed with 0.1% fast green dye (Sigma-Aldrich). The final concentration for each plasmid was ~2.5 μg/μL. Dams were anesthetized with an intraperitoneal injection of ketamine/xylazine cocktail and a midline laparotomy was performed to expose the uterine horns. A pulled glass beveled micropipette was used to inject 1–2 μL of the plasmid DNA through the uterine wall and amniotic sac into the lateral ventricle of the mouse embryo. The anode of a Tweezertrode (Harvard Apparatus) was placed over the dorsal surface of the parietal cortex above the uterine wall and four 35 V pulses of 50 ms duration were applied, separated by a 950 ms interval with a BTX ECM830 pulse generator (Harvard Apparatus). After the electroporation, the abdominal cavity was filled with warm 0.9% saline and the incision closed with absorbable sutures. Dams were placed in a clean cage and closely monitored during recovery. All procedures were reviewed and approved by the Institutional Animal Care and Use Committee at Boston University School of Medicine. Experimental Subjects At birth, mice were screened for electroporation and housed with their mothers. A total of 52 mice, both male and female, were used in this study as follows: n = 23 at P21 for electrophysiological recordings and cell morphology studies; n = 4 at P21 for stereology; n = 12 total for time course experiments (n = 3 each at E16.5, P0, P3, and P7); n = 4 at E14.5 for Tbr2 immunohistochemistry; n = 3 at E14.5 for Tbr2 in situ hybridization; n = 3 at E14.5 for Tbr2-Cre × Tbr2-SL at 30 h post-IUE; n = 3 at E15 for Tbr2-Cre × Tbr2-SL at 40 h post-IUE. All animals were handled according to animal care guidelines from the NIH “Guide for the Care and Use of Laboratory Animals” and “the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals” and research procedures were approved by the Institutional Animal Care and Use Committee at Boston University School of Medicine. Tissue Processing and Confocal Imaging For the determination of number and depth of Tbr2 and non-Tbr2-derived cells in layer 4 of the barrel cortex, electroporated mice were anesthetized with an intraperitoneal injection of ketamine/xylazine at postnatal day 21 and intracardially perfused with 4% paraformaldehyde (PFA). For the time course study, the brains of electroporated mice at each age (E14.5, E16.5, P0 and P3, P7) were collected and fixed overnight in 4% PFA. All brains were cryoprotected by submersion in 15% sucrose in 0.01 M PBS for 24 h, followed by 30% sucrose for 24 h. After cryoprotection the brains were frozen in Optimal Cutting Temperature (OCT) Compound and stored at −80 °C. About Twenty micrometre or 50 μm-thick serial sections were cut in the coronal plane throughout the rostrocaudal extent of the barrel field (between bregma 0.50 mm and bregma −1.94 mm) (Paxinos, 2003) using a cryostat (Microm HM 560). The 20 μm sections were directly mounted on superfrost slides and air-dried for 20 min. The 50 μm sections were stored in anti-freeze solution (30% ethylene glycol, 30% glycerol in 0.05 M PB) at −20 °C. For stereology every fifth 50 μm section was collected for subsequent immunohistochemistry. The first section was chosen randomly and a total of 7–8 sections per brain were immunostained for VGLUT2 to label layer 4 of the barrel field. Tile scans of the entire electroporated barrel field were acquired at 0.208 × 0.208 × 1 μm voxel resolution with a Plan-Apochromat 20×/0.8 NA objective lens and 1.5× digital zoom using an upright Zeiss LSM 710 microscope with a motorized stage and the Zen MultiTime macro. Stereological Counting Methods The virtual sections were then uploaded to StereoInvestigator software (MBF Bioscience, Williston, VT) for counting using the optical fractionator procedure to generate unbiased counting frames (West et al., 1991; Peterson, 1999). The numbers of Tbr2 and non-Tbr2 neurons in both the barrel core and septum were determined using the following stereological parameters: sampling grid 100 × 100 μm, counting frame 100 × 100 μm, optical dissector height 18 μm (14 μm plus 2μm guard zones on either side). A total of 6000 neurons were counted. The wall/core ratio was calculated by dividing the number of neuronal somata in the wall by the number of neurons in the core for each group. The files generated with Stereoinvestigator were imported to Neurolucida Explorer software (MBF Bioscience), and the depth of each Tbr2 and non-Tbr2 cell was measured as the distance from the corresponding marker to the border of layers 4 and 5, on the basis of the VGLUT2 immunostain. The Nearest Neighbor Distance (NND) in this dataset was assessed with the NND tool of the Neurolucida Explorer software. The NND of each neuron was divided by average NNDs of all neurons in each animal for normalization. For the time course study, one representative section of each brain was scanned at 0.208 × 0.208 × 1 μm voxel resolution, the neurons counted, and their depth relative to the pia measured using Neurolucida software. Immunohistochemistry Free floating 50 µm sections or 300 µm slices (from which recordings were obtained) were rinsed in 0.01 M PBS and antigen retrieval was performed using 10 mM Sodium Citrate Buffer (pH 8.5) for 30 min in a 60° water bath. Binding sites were unmasked by incubation in 50 mM glycine for 1 h. Tissue was then incubated in blocking solution (10% BSA, 0.2% triton-x in 0.01 M PBS) for 1 h. After blocking, tissue was incubated in primary antibody for 1 day (for 50 μm sections) or for 7 days (for 300 μm slices). After rinsing thoroughly with 0.01 M PBS, 300 μm slices were incubated in secondary antibody for 48 h. Fifty micrometre sections were incubated in biotinylated secondary antibodies for 2 h, followed by incubation with Streptavidin Alexa Fluor®-633 for 24 h. Primary and secondary antibodies were diluted in 0.2% BSAc, 0.1% triton-x, 0.1 M phosphate buffer and their penetration enhanced by controlled microwaving (150 W for 10 min at 35 °C). Finally, tissue was rinsed, mounted in Prolong anti-fade medium (Invitrogen) and coverslipped. For immunohistochemistry on slide mounted 20 μm sections, sections were rinsed in 0.01 M PBS and then placed in 10 mM Sodium Citrate Buffer (pH 8.5) for antigen retrieval by microwaving at 800 W for 1 min and then at 80 W for 10 min. Sections were subsequently incubated in blocking solution for 1 h and then in primary antibody overnight. After rinsing thoroughly with 0.01 M PBS, sections were incubated in secondary antibody for 2 h. Finally, sections were rinsed, and coverslipped in DAPI-containing Vectashield mounting medium. The following pairs of primary/secondary antibodies were used: 1) Anti-Tbr2 (polyclonal rabbit, 1:200, AB2283, Millipore)/633 goat anti-rabbit, 1:200, A21070, ThermoFisher Scientific; 2) Anti-vesicular glutamate transporter 2 (anti-VGLUT2 polyclonal guinea pig, 1:1000, AB2251, Millipore)/biotinylated goat anti-guinea pig IgG (1:200, BA-7000, Vector Laboratories) and Streptavidin Alexa Fluor®-633 (1:400, S21375, ThermoFisher Scientific) for 20 and 50 μm sections or secondary goat anti-guinea pig IgG conjugated with Alexa Fluor®-647 (1:200, 706-605-148, Jackson Immunoresearch) for 300 μm slices; 3) Anti-ZsGreen (polyclonal rabbit, 1:500, 632 474, Living Colors®)/488 goat anti-rabbit (1:200, A11070, ThermoFisher Scientific); 4) Anti-mCherry (monoclonal mouse, 1:1000, 632 543, Living Colors®)/546 goat anti-mouse (1:200, A11018, ThermoFisher Scientific). Fluorescence In Situ Hybridization and Immunohistochemistry Slide mounted 20 μm sections were fixed in 4% formaldehyde in PBS for 5 min. Sections were washed with DEPC-MQ and acetylated in 0.25% Acetic anhydrite in 0.1 M TEA. Sections were then incubated in pre-hybridization buffer (50% formamide, 5 × SSPE, 0.1% SDS, 0.64 mg/mL Yeast tRNA) for 60 min at room temperature, followed by incubation with digoxigenin-labeled antisense RNA probe (500 ng/mL of Tbr2 probe in pre-hybridization buffer) at 60 °C for 15–20 h. Hybridized sections were washed with 2 × SSC in 50% formamide at 60 °C, 0.2 × SSC in 50% formamide at 42 °C and 0.2 × SSC in MQ at room temperature for 20 min each and rinsed with MABT. To block the endogenous peroxidase, sections were incubated in 0.3% hydrogen peroxide and then washed with MABT. Sections were then incubated in 2% blocking reagent (Roche) in MABT for 60 min and then in POD-conjugated anti-digoxigenin antibody (Roche, diluted 1:100) for 60 min. After washing with MABT, fluorescence deposition was performed using TSA plus Cyanine 3 system (Perkin Elmer, diluted 1:50) for 60 min. Sections were washed with MABT and PBS, incubated in primary antibodies (rabbit anti-ZsGreen, 1:500, 632 474, Living Colors; rat anti-RFP, 1:200, 5F8, Chromo Tek) overnight. After washing with 0.01% Triton X-100 in PBS, sections were incubated in Alexa fluor 488 and 568 conjugated secondary antibodies (1:250, A11070, A11077, Life Technologies) for 3 h. Sections were washed with 0.01% Triton X-100 in PBS and then incubated in Hoechst 33 342 in PBS (diluted 1:1000, Invitrogen) for 3 min. After washing with PBS, in situ hybridization and immunostained sections were mounted in Fluoromount-G (Southern Biotech) and imaged on a laser-scanning confocal microscope (LSM 710, Carl Zeiss). Electrophysiology At postnatal day 21, electroporated mice were sedated with isoflurane and decapitated. Brains were extracted and placed in oxygenated (95% O2/5% CO2) ice-cold Ringer’s solution (concentrations in mM: 26 NaHCO3, 124 NaCl, 2 KCl, 3 KH2PO4, 10 glucose, 2.5 CaCl2, and 1.3 MgCl2, pH 7.4, Sigma-Aldrich). Coronal slices of 300 μm thickness through the parietal cortex were cut using a vibrating microtome and equilibrated for 1 h in oxygenated Ringer’s solution at room temperature. Individual slices were then placed in a submersion type recording chamber (Harvard Apparatus) on the stage of Nikon E600 IR-DIC microscopes (Micro Video Instruments) and continually perfused with room temperature oxygenated Ringers solution at 2.5 mL/min. Barrels in layer 4 of the somatosensory cortex were visualized under IR-DIC optics and mCherry+ and ZsGreen+ neurons were identified under epifluorescence. Whole-cell patch clamp recordings were performed as previously described (Rocher et al. 2010; Crimins et al. 2012; Amatrudo et al. 2012), using electrode pipettes made from non-heparinized microhematocrit capillary tubes (Sutter Instrument) with a Flaming and Brown horizontal pipette puller (model P87, Sutter Instrument). Patch electrode pipettes were filled with potassium gluconate (KGlu) internal solution (concentration in mM: 122 KGlu, 2 MgCl2, 5 EGTA, and 10 NaHEPES containing 1% biocytin pH 7.4; Sigma-Aldrich) and had resistances between 3 and 7 MΩ. Data were acquired with EPC9 or EPC10 amplifiers and PatchMaster software (HEKA Elektronik). Access resistance was monitored throughout the experiment and all recordings were low-pass filtered at 10 kHz. A total of 23 mice, male and female, were used for these experiments. The total number of lineage-identified neurons recorded from was: n = 31 (Tbr2-derived neurons) and n = 24 (non-Tbr2-derived neurons). Intrinsic Membrane and Action Potential Properties Whole-cell patch clamp recordings in the current-clamp mode were used to assess passive membrane properties and action potential (AP) firing properties. Passive membrane properties included resting membrane potential (Vr), input resistance (Rn) and membrane time constant. Vr was measured as the voltage in the absence of current injection. A series of 200 ms hyperpolarizing and depolarizing current steps was applied for the rest of the measures. The voltage responses to each step were measured at steady state and plotted on a voltage-current graph and Rn was calculated as the slope of the best-fit line through the linear portion of the plot. Membrane time constant was measured by fitting a single-exponential function to the membrane voltage response to the −10 pA hyperpolarizing step. Rheobase, the amount of current needed to elicit the first AP, was determined with a 10 s depolarizing current ramp (0–200 pA; 3.03 kHz sampling frequency). Single AP properties, including threshold and amplitude, were measured on the first evoked AP in a 200 ms current-clamp series using an expanded timescale and the linear measure tool in FitMaster analysis software (HEKA Elektronik). Threshold was determined as the sharp, upward deflection in the voltage trace. Amplitude and rise time were measured from the voltage threshold to the peak of the AP. Finally, a series of 2 s hyperpolarizing and depolarizing steps (−200 to +450 pA, using 25 or 50 pA increments, 12.5 kHz sampling frequency) was used to assess repetitive AP firing. Firing rates at each current step were subsequently calculated. At the −200 pA hyperpolarizing step, the FitMaster linear measure tool was used to measure the amplitude of the sag potential (a depolarizing H-current mediated response). Spontaneous Excitatory Postsynaptic Currents, sEPSCs AMPA receptor-mediated spontaneous excitatory currents (sEPSCs) were recorded for 2 min at a holding potential of −80 mV (6.67 kHz sampling frequency). Minianalysis software (Synaptosoft) was used to assess synaptic current properties including: frequency, amplitude, area, time to rise and time to decay. For assessment of kinetics, the rise and decay of averaged traces were each fit to a single-exponential function. For all synaptic current measurements, the event detection threshold was set at the maximum root mean squared noise level (5 pA). Cell Morphometry During the recordings, neurons were simultaneously filled with 1% biocytin in the internal solution (pH 7.4; Sigma-Aldrich). Following recordings, slices containing filled neurons were fixed in 4% paraformaldehyde in 0.1 M PBS solution (pH 7.4) for 2 days. After rinsing, slices were incubated in Streptavidin-Alexa 405 (1:500) for 2 days. Lineage identity was confirmed using confocal microscopy by acquiring z-stacks through the cell body at 20× (3× digital zoom) at excitation wavelengths of 488 and 594 nm for ZsGreen and mCherry, respectively. Following lineage identification, slices that contained filled neurons were processed with biotinylated goat anti-Streptavidin (1:400, BA0500, Vector Labs) followed by Alexa-Streptavin-488 (to amplify the signal of the filled cell), and then immunostained with VGLUT2 to identify the position of the filled cell relative to the barrel field (immunohistochemistry procedures described above). Well-filled Layer 4 (L4) neurons located in the barrel field were then imaged in their entirety using a Plan Apochromatic 20×/0.8 NA DIC air-immersion objective lens. For 3D reconstruction, optical sections were acquired at a resolution of 0.2 × 0.2 × 0.4 μm per voxel. Z-stacks of collected TIFF images were deconvolved (AutoDeblur software, Media Cybernetics) to reduce z-plane signal blurring. Z-stacks for each neuron were imported to Neurolucida software (MBF Bioscience) for 3D alignment and reconstruction. The interactive semi-automated tool was used for reconstruction of the soma and the entire dendritic tree, and edited manually. The generated DAT file was uploaded to Neurolucida Explorer software to measure dendritic length, dendritic complexity, and Sholl analysis. The dendritic complexity index was calculated from: (sum of the terminal orders + number of terminals) * (total dendritic length/number of primary dendrites). The polarization of the dendritic tree was calculated by obtaining a round histogram where length was plotted as a function of direction (from 0 to 360o), each bin was then normalized to the total dendritic length and a vector was calculated as the sum of all bins and their corresponding orientation. The absolute value of the X component of the vector was used as a quantitative measure of horizontal polarization (or asymmetry), and the value of the Y component as a measure of vertical polarization. The total numbers of neurons reconstructed and analyzed were: non-Tbr2 pyramidal n = 9, non-Tbr2 spiny stellate n = 13, Tbr2 pyramidal n = 13, Tbr2 spiny stellate n = 16. Spine Density and VGLUT2+ Appositions For each cell, 1–2 apical and basal branches were selected and imaged in their entirety using a Plan Apochromatic 63×/1.3 NA DIC oil-immersion objective lens and 1.5 digital zoom at wavelengths of 488 and 633 nm, for the filled cell and VGLUT2, respectively. Image stacks were acquired at a resolution of 0.044 × 0.044 × 0.2 μm per voxel. Z-stacks were deconvolved and imported to Neurolucida for alignment and semi-automated reconstruction. Spines were manually identified and marked in Neurolucida software, the diameter of their heads measured, and classified into subtypes as follows: spines without a neck were classified as stubby, spines with a neck and with a head diameter of less than 0.6 μm were classified as thin, spines with a head diameter of 0.6 μm or more were classified as mushroom, and spines with necks longer than 3 μm were classified as filopodia. For appositions, VGLUT2-immunolabeled boutons that were apposed to spines of filled cells were manually marked and classified by spine type. To be considered an apposition, the overlap was required to occur between the highest intensity points of the fluorescent label of both structures (spine and bouton) and to be present in at least 3 z-steps. The DAT file was imported into Neurolucida Explorer to measure spine number by subtype, dendritic length, apposition number by class, and to run Sholl analyses for the distribution of spines and appositions relative to distance from the soma. Spine and VGLUT2 appositions were quantified for the following n of neurons: non-Tbr2 pyramidal n = 4, non-Tbr2 spiny stellate n = 4, Tbr2 pyramidal n = 5, Tbr2 spiny stellate n = 5. Statistical Analyses Differences between the 2 lineages were assessed using a Student’s t-test and error bars represent SEM. A coefficient of error (CE, Gundersen m = 1) was used in the stereology study to measure the precision of stereological estimates (Gundersen and Jensen, 1987) and was considered appropriate below 0.10. For the numbers and locations of cells, a negative binomial model was employed with generalized estimating equations (GEE). For the depth comparison, a mixed linear model that used a covariance adjustment of compound symmetry was run to account for variability across animals. Cumulative frequency distributions of normalized NNDs of neurons from each lineage were compared using 2-sample Kolmogorov–Smirnov test in MATLAB (R2015, Mathworks). Finally, a k-means cluster analysis was run to group all cells as either stellate or pyramidal cells. In all cases, differences were considered significant if the P-value was <0.05. Results In Utero Fate-mapping at E13.5 Labels Progenitors that Generate Layer 4 Neurons of the Barrel Cortex We took advantage of the unique pattern of transcription factor expression to indelibly label and separate Tbr2 expressing progenitors from Tbr2 negative progenitors, as well as their daughter cells (Tyler et al. 2015) (Fig. 1A). To do so we paired a Tbr2-Cre driver with a CAG-lox-ZsGreen-lox-mCherry (CAG-SL) reporter plasmid. In the absence of Cre, ZsGreen was expressed in all precursors and their neuronal progeny. In cells where the Tbr2 promoter was active, Cre mediated recombination of the ZsGreen cassette led to the expression of mCherry. Thus, our 2-color genetic fate-mapping technique labeled the neuronal progeny of non-Tbr2 precursors green, while neurons derived from Tbr2 precursors were labeled red (Fig. 1A). Not surprisingly due to the temporal lag between fluorescent protein expression and Cre mediated recombination, many cells expressed both ZsGreen and mCherry. Accordingly, these cells were assigned to the Tbr2 lineage as any amount of mCherry protein was indicative of pTbr2 expression. To validate our fate-mapping strategy, we quantified the numbers of electroporated cells within 100 μm of the ventricular surface that expressed mCherry, ZsGreen, and Tbr2 protein at embryonic day E14.5, 24 h post in utero electroporation (IUE) (Fig. 1B,C). At this time point, 59.6% of the cells expressed mCherry and the majority was a group of bIPCs that expressed Tbr2 protein (40.9%). We also observed that 18.7% of the IUE cells expressed mCherry, but did not stain positive for Tbr2. These cells may represent newly generated bIPCs in which the Tbr2 promoter is active, but a detectable level of Tbr2 protein has not yet accumulated (Tyler et al. 2015; Vasistha et al. 2015), or they may be neurons. To test this hypothesis, we performed in situ hybridization to detect Tbr2 mRNA in the electroporated cells, 24 h after surgery (Fig. 1D). We found that 94.7% of the cells that expressed mCherry also expressed Tbr2 mRNA (Fig. 1E). The few red cells that did not show Tbr2 mRNA expression were located away from the ventricular surface and thus were likely neurons born from the Tbr2 lineage during the post-IUE interval. Apical progenitors comprised the second largest group of IUE cells (31.4%) and expressed ZsGreen only. Importantly, we detected a small number of cells distributed throughout the electroporated field that expressed detectable levels of Tbr2 protein or mRNA, (9% and 13.2% of total IUE cells, respectively), but not mCherry (Fig. 1C,E). Because this may reflect insufficient recombination, and thus, constitute a caveat of our study, we designed a strategy to measure the recombination efficiency of the Tbr2-Cre plasmid. To do so, we used the combination of Tbr2-Cre and Tbr2-SL (Fig. 1F). With the Tbr2-SL plasmid, fluorescent proteins are only produced in cells when the Tbr2 promoter is active and mCherry serves as a direct readout of recombination efficiency driven by the Tbr2-Cre plasmid. Thus, in the case of 100% co-electroporation of Tbr2-Cre and Tbr2-SL, all transfected cells should express mCherry if recombination is 100% efficient. Analysis 30 h post-IUE with these plasmids, revealed 88% of cells expressed mCherry (Fig. 1F, right panel), a result similar to the experiment performed with the Tbr2-Cre × Cag-SL combination at E13.5–E14.5. Importantly however, at 40 h post-IUE, recombination occurred in virtually all the cells (97.8%; Fig. 1F). These results demonstrate that our labeling paradigm closely approximates the endogenous expression of Tbr2 at E13.5 and that our combination of plasmids efficiently separates Tbr2 and non-Tbr2 lineages. Figure 1. View largeDownload slide Genetic fate-mapping at E13.5 separately labels Tbr2 and non-Tbr2 neocortical progenitor lineages. (A) A pTbr2-cre plasmid was co-transfected with a CAG-stoplight reporter plasmid using in utero electroporation (IUE) at E13.5 to fate map Tbr2 lineage cells with mCherry and non-Tbr2 cells with ZsGreen only. VZ: ventricular zone; SVZ: subventricular zone. (B) Confocal micrograph of a coronal section through the embryonic cortex 24 h after IUE, showing Tbr2 lineage cells labeled with mCherry and non-Tbr2 lineage cells labeled with ZsGreen only (left and right panels) and Tbr2 protein expressing cells in blue (right panel). Scale bar, 20 μm. (C) Bar graphs of mCherry+/Tbr2 immunonegative cells (dark red), mCherry+/Tbr2 immunopositive (light red), ZsGreen+/Tbr2 immunonegative cells (light green) and ZsGreen+/Tbr2 immunopositive cells (dark green), expressed as percentage of total electroporated cells. (D) Confocal micrograph of a coronal section through the embryonic cortex 24 h after IUE, showing Tbr2 lineage cells labeled with mCherry, non-Tbr2 lineage cells labeled with ZsGreen only and Tbr2 mRNA in white. Scale bar, 20 μm. (E) Bar graphs of mCherry+/Tbr2 mRNA negative cells (dark red), mCherry+/Tbr2 mRNA positive (light red), ZsGreen+/Tbr2 mRNA negative cells (light green) and ZsGreen+/Tbr2 mRNA positive cells (dark green), expressed as percentage of total electroporated cells. (F) Left panel: Confocal micrograph at 40 h post-IUE, showing in red Tbr2 lineage cells where recombination occurred, and in green Tbr2 lineage neurons where no recombination has taken place. IZ: intermediate zone; scale bar, 50 μm; n = 4. Right panel: bar graphs of non-recombined Tbr2 cells (ZsGreen+/mCherry−) and recombined Tbr2 cells (mCherry+) at 30 h and at 40 h post-IUE. Figure 1. View largeDownload slide Genetic fate-mapping at E13.5 separately labels Tbr2 and non-Tbr2 neocortical progenitor lineages. (A) A pTbr2-cre plasmid was co-transfected with a CAG-stoplight reporter plasmid using in utero electroporation (IUE) at E13.5 to fate map Tbr2 lineage cells with mCherry and non-Tbr2 cells with ZsGreen only. VZ: ventricular zone; SVZ: subventricular zone. (B) Confocal micrograph of a coronal section through the embryonic cortex 24 h after IUE, showing Tbr2 lineage cells labeled with mCherry and non-Tbr2 lineage cells labeled with ZsGreen only (left and right panels) and Tbr2 protein expressing cells in blue (right panel). Scale bar, 20 μm. (C) Bar graphs of mCherry+/Tbr2 immunonegative cells (dark red), mCherry+/Tbr2 immunopositive (light red), ZsGreen+/Tbr2 immunonegative cells (light green) and ZsGreen+/Tbr2 immunopositive cells (dark green), expressed as percentage of total electroporated cells. (D) Confocal micrograph of a coronal section through the embryonic cortex 24 h after IUE, showing Tbr2 lineage cells labeled with mCherry, non-Tbr2 lineage cells labeled with ZsGreen only and Tbr2 mRNA in white. Scale bar, 20 μm. (E) Bar graphs of mCherry+/Tbr2 mRNA negative cells (dark red), mCherry+/Tbr2 mRNA positive (light red), ZsGreen+/Tbr2 mRNA negative cells (light green) and ZsGreen+/Tbr2 mRNA positive cells (dark green), expressed as percentage of total electroporated cells. (F) Left panel: Confocal micrograph at 40 h post-IUE, showing in red Tbr2 lineage cells where recombination occurred, and in green Tbr2 lineage neurons where no recombination has taken place. IZ: intermediate zone; scale bar, 50 μm; n = 4. Right panel: bar graphs of non-recombined Tbr2 cells (ZsGreen+/mCherry−) and recombined Tbr2 cells (mCherry+) at 30 h and at 40 h post-IUE. Differential Localization of Tbr2 and Non-Tbr2 Lineage Neuronal Somata within Layer 4 of the Barrel Cortex IUE enabled us to label neurons generated within a narrow time frame, as the plasmids introduced by electroporation remain episomal (Stancik et al. 2010). Specifically, IUE at E13.5 labeled neurons primarily destined for layer 4, which we were able to demarcate using immunohistochemistry for Vesicular Glutamate Transporter 2 (VGLUT2; Fig. 2A) which predominantly labels thalamocortical nerve terminals ending in this layer. Analysis of P21 brains revealed that the fate-mapped neurons spanned the full extent of layer 4, with the majority of cells located within the VGLUT2-positive region, between 400 and 650 μm deep to the pia (30–50% of cortical depth, Fig. 2B), although there were some labeled neurons in layers 2–3 and layer 5. Of all the electroporated cells, 62.3% expressed mCherry and were derived from the Tbr2 lineage while 37.7% were non-Tbr2 lineage neurons and expressed ZsGreen alone. Interestingly, although both Tbr2 and non-Tbr2-derived neurons were distributed across the full extent of layer 4, Tbr2-derived neurons were on average positioned deeper within layer 4 (distance from the layer 5 to layer 4 boundary, non-Tbr2 = 122.44±8.65 μm; Tbr2 = 108.49±8.63 μm, P < 0.01) (Fig. 2C). Figure 2. View largeDownload slide Tbr2 and non-Tbr2 lineage-derived neurons that co-populate layer 4 of the barrel cortex differ in number and distribution. (A) Coronal section of the somatosensory cortex at P21 showing VGLUT2+ barrels (cyan) and electroporated cells. Scale bar, 100 μm. (B) Distribution of non-Tbr2 and Tbr2 lineage neuronal cell bodies (represented by open and closed circles, respectively) throughout the normalized cortical depth at P21, dashed lines indicate the upper and lower limits of layer 4, green and red lines indicate mean depths for non-Tbr2 and Tbr2 lineage populations, respectively. (C) Mean distance of Tbr2 lineage neurons versus non-Tbr2 lineage neurons from the superficial limit of layer 5 (the lowest limit of layer 4) **P < 0.01. (D) Coronal sections through the somatosensory cortex showing electroporated neurons (red and green) and their location relative to VGLUT2+ thalamic terminals (cyan), from E16.5 to P7. Scale bar, 100 μm. CP: cortical plate; SP: subplate; VZ: ventricular zone; SVZ: subventricular zone; WM: white matter; cortical Layers: L1, L2–3, L4, L5, L6. (E) Changes in average depth of Tbr2 and non-Tbr2 lineage neurons from E16.5 to P21, and their position relative to thalamocortical afferents (blue shading). (E14.5 n = 4, E16.5 n = 4, P0 n = 3, P3 n = 3, P7 n = 2, P21 n = 4). (F) Normalized average nearest neighbor distance (NND; top panel), and cumulative frequency distribution of NND (bottom panel, dark lines and lighter shaded areas indicate mean and SEM, respectively) of Tbr2 and non-Tbr2 lineage neurons in layer 4. Top panel ***P < 0.001 (Student’s t-test), bottom panel *P < 0.01 (2-sample Kolmogorov–Smirnov test). (G) Top panel—representative 2D projection of the distribution of electroporated cells (red and green dots) within VGLUT2+ barrel cores (black arrows) and VGLUT2− barrel walls (black arrowheads, pink outlines) across the barrel field. Bottom panel—proportion of Tbr2 and non-Tbr2 lineage neurons in the barrel field, in total and by compartment. (H) Mean Barrel Wall/Core location ratios showing differences between the 2 lineages ***P < 0.0001. Figure 2. View largeDownload slide Tbr2 and non-Tbr2 lineage-derived neurons that co-populate layer 4 of the barrel cortex differ in number and distribution. (A) Coronal section of the somatosensory cortex at P21 showing VGLUT2+ barrels (cyan) and electroporated cells. Scale bar, 100 μm. (B) Distribution of non-Tbr2 and Tbr2 lineage neuronal cell bodies (represented by open and closed circles, respectively) throughout the normalized cortical depth at P21, dashed lines indicate the upper and lower limits of layer 4, green and red lines indicate mean depths for non-Tbr2 and Tbr2 lineage populations, respectively. (C) Mean distance of Tbr2 lineage neurons versus non-Tbr2 lineage neurons from the superficial limit of layer 5 (the lowest limit of layer 4) **P < 0.01. (D) Coronal sections through the somatosensory cortex showing electroporated neurons (red and green) and their location relative to VGLUT2+ thalamic terminals (cyan), from E16.5 to P7. Scale bar, 100 μm. CP: cortical plate; SP: subplate; VZ: ventricular zone; SVZ: subventricular zone; WM: white matter; cortical Layers: L1, L2–3, L4, L5, L6. (E) Changes in average depth of Tbr2 and non-Tbr2 lineage neurons from E16.5 to P21, and their position relative to thalamocortical afferents (blue shading). (E14.5 n = 4, E16.5 n = 4, P0 n = 3, P3 n = 3, P7 n = 2, P21 n = 4). (F) Normalized average nearest neighbor distance (NND; top panel), and cumulative frequency distribution of NND (bottom panel, dark lines and lighter shaded areas indicate mean and SEM, respectively) of Tbr2 and non-Tbr2 lineage neurons in layer 4. Top panel ***P < 0.001 (Student’s t-test), bottom panel *P < 0.01 (2-sample Kolmogorov–Smirnov test). (G) Top panel—representative 2D projection of the distribution of electroporated cells (red and green dots) within VGLUT2+ barrel cores (black arrows) and VGLUT2− barrel walls (black arrowheads, pink outlines) across the barrel field. Bottom panel—proportion of Tbr2 and non-Tbr2 lineage neurons in the barrel field, in total and by compartment. (H) Mean Barrel Wall/Core location ratios showing differences between the 2 lineages ***P < 0.0001. To determine how this differential localization arises during development, we assessed changes in the radial position of E13.5 fate-mapped neurons and their relationship to ingrowing thalamocortical afferents (TCAs) between E16.5 and P21 (Fig. 2D,E). At E16.5, most of the fate-mapped neurons were located in the cortical plate (CP), although some were still migrating from the VZ and SVZ. During this period of development, thalamic axons are “waiting” in the subplate (Lopez-Bendito and Molnar 2003; Li and Crair 2011) and have not yet invaded the cortex, as shown by VGLUT2 staining. At P0, almost all of the electroporated cells were located in superficial layers and the thalamocortical axons had arrived at the CP. By P3, a distinct band of thalamic afferents delimited layer 4, and the majority of the fate-mapped cells were found intermingled with TCAs, although the barrels were still not apparent. By P7, the barrels had formed and neurons were located primarily in layer 4. Importantly, we first detected a difference in the positioning of Tbr2-derived neurons compared to non-Tbr2-derived neurons at P7, a difference that became more prominent at P21 (Fig. 2E). These results demonstrate that the majority of neurons labeled by IUE at E13.5 are derived from Tbr2 expressing precursors, populate layer 4, and that their final position within the layer is not established until after the first postnatal week. To determine whether cells derived from the 2 lineages show different degrees of spatial organization, we performed high-resolution stereological analysis of Tbr2 and non-Tbr2-derived neurons within layer 4 followed by nearest neighbor distance (NND) analyses. We found that layer 4 Tbr2 lineage neurons clustered to a greater extent than non-Tbr2 lineage neurons (Fig. 2F). In layer 4 of the rodent somatosensory cortex, neurons are spatially allocated into compartments that form columnar barrel cores and interstitial zones that form barrel walls and septa (Woolsey and Van der Loos 1970). Therefore, we assessed the location of Tbr2 and non-Tbr2 lineages neurons within each compartment to determine whether the clustering we observed corresponded to a differential allocation of cells between the VGLUT2− barrel wall and VGLUT2+ barrel core (Fig. 2G). Within the barrel field as a whole, 56.8% of the electroporated neurons were derived from the Tbr2 lineage while 43.2% were derived from the non-Tbr2 lineage. In the barrel core, 55.1% of the neurons were Tbr2 lineage while 44.9% were non-Tbr2 neurons. In the barrel wall, 60.1% of the neurons were Tbr2 lineage, while 39.9% were of the non-Tbr2 lineage. Thus, Tbr2-derived neurons preferentially localized to the barrel wall over the barrel core, when compared to non-Tbr2 cells. The ratio of number of cells in the barrel wall to number of cells in the barrel core was 1.12 for Tbr2 cells and 0.89 for non-Tbr2 cells (P < 0.0001) (Fig. 2H). Tbr2 and Non-Tbr2 Lineage Neurons Generate both Spiny Stellate and Pyramidal Neurons In vitro whole-cell patch clamp recordings and intracellular filling techniques were employed to compare the detailed morphological and electrophysiological properties of Tbr2 (n = 31) and non-Tbr2 (n = 24) derived neurons. In our random sample, 57% of the recorded and filled cells were derived from Tbr2 progenitors, while 43% were non-Tbr2 lineage neurons (Fig. 3). This is consistent with the stereology results (Fig. 2G), suggesting that the neurons sampled were representative of the overall population. Figure 3. View largeDownload slide Tbr2 and non-Tbr2 progenitors give rise to both spiny stellate and pyramidal neurons in layer 4 of the barrel cortex. (A) Filled neurons were imaged for lineage identification and their location relative to the barrel field determined on the basis of VGLUT2 staining (dashed lines indicate barrels). Parameters quantified are indicated and included: distance (D) to layer 5 (L5); L4 thickness; apical dendrite (AD) span; soma-to-pia distance. Cortical Layers: L1, L2–3, L4, L5. Scale bar, 200 μm. (B) Reconstructions of representative Tbr2 and non-Tbr2 lineage neurons intracellularly filled during whole-cell patch clamp recordings; apical dendrites highlighted in green (non-Tbr2) and red (Tbr2). Scale bar, 200 μm. (C) Scatter plot of soma position versus AD% (AD extent as a percentage of soma-to-pia distance) separated the neurons into “Spiny stellate” and “Pyramidal” groups (with a k-means cluster analysis; P < 0.0001). White arrow indicates position and AD of neuron indicated in (A). (D) Proportions of spiny stellate and pyramidal neurons in L4 of the barrel cortex that originate from Tbr2 and non-Tbr2 progenitors. Figure 3. View largeDownload slide Tbr2 and non-Tbr2 progenitors give rise to both spiny stellate and pyramidal neurons in layer 4 of the barrel cortex. (A) Filled neurons were imaged for lineage identification and their location relative to the barrel field determined on the basis of VGLUT2 staining (dashed lines indicate barrels). Parameters quantified are indicated and included: distance (D) to layer 5 (L5); L4 thickness; apical dendrite (AD) span; soma-to-pia distance. Cortical Layers: L1, L2–3, L4, L5. Scale bar, 200 μm. (B) Reconstructions of representative Tbr2 and non-Tbr2 lineage neurons intracellularly filled during whole-cell patch clamp recordings; apical dendrites highlighted in green (non-Tbr2) and red (Tbr2). Scale bar, 200 μm. (C) Scatter plot of soma position versus AD% (AD extent as a percentage of soma-to-pia distance) separated the neurons into “Spiny stellate” and “Pyramidal” groups (with a k-means cluster analysis; P < 0.0001). White arrow indicates position and AD of neuron indicated in (A). (D) Proportions of spiny stellate and pyramidal neurons in L4 of the barrel cortex that originate from Tbr2 and non-Tbr2 progenitors. Excitatory neurons in layer 4 can be classified into 3 main morphological types: spiny stellate neurons, which lack a prominent apical dendrite; pyramidal neurons, which possess an apical dendrite with a sparse apical tuft; and star pyramids which extend a primary apical dendrite without a tuft (Jones 1975; Staiger et al. 2004; Oberlaender et al. 2012). Importantly, the prominent apical dendrite of pyramidal neurons enables them to sample heterogeneous inputs to superficial layers 1–3. Conversely, spiny stellate neurons achieve their distinct star-shaped morphology via selective regression of the apical dendrite as they begin to specialize in the reception of thalamic inputs. Although these classes of neurons are distinguished based on the prominence and extent of their apical dendrite, they represent a developmental continuum from pyramidal to stellate morphology (Lund 1984; Callaway and Borrell 2011; Vercelli et al. 1992; Supplementary (S) Fig. S1). To classify the neurons in our sample in an unbiased way, we measured the extent of the apical dendrite (identified in all cells as the largest caliber dendrite running perpendicular to the pial surface) as the percentage of the distance from the soma to the pia (AD%) (Fig. 3A), as previously described (Callaway and Borrell 2011). Neurons classified as pyramidal had an AD% close to 100, with an apical dendrite that extended to layer 1, while at the other end of the continuum, spiny stellate neurons had a low AD% since their apical dendrite remained restricted to layer 4 (Fig. 3B). A k-means clustering analysis based on AD% separated the neurons in our sample into 2 groups: 1 with an AD% ranging from 0% to 41% (the spiny stellate group) and 1 ranging from 56% to 100% (the pyramidal group, which included both star pyramids and pyramidal neurons) (P < 1E−26; Fig. 3C). Tbr2 and non-Tbr2 lineages both generated spiny stellate and pyramidal neurons, with the Tbr2 lineage producing a larger proportion of each morphological subtype (Fig. 3D). Spiny Stellate and Pyramidal Neurons Possess Distinct Morphological Features based on Neural Precursor Lineage To determine whether lineage influenced the morphological features of layer 4 neurons, we imaged the intracellularly filled cells using high-resolution confocal microscopy and assessed their dendritic topology (Fig. 4; Supplementary Table S1). Examples of representative filled and reconstructed spiny stellate and pyramidal neurons are shown in Figures 4A–D and S1. Two-dimensional projections of convex hull volumes of apical and basal dendritic arbors for reconstructed Tbr2 (red) and non-Tbr2 (green) neurons are indicated with dark and light shading, respectively (Fig. 4B,D). Figure 4. View largeDownload slide Spiny stellate and pyramidal neurons have different structural properties depending on their lineage. (A) Confocal image of a representative spiny stellate neuron. Scale bar, 50 μm. (B) Two-dimensional projections of convex hull volumes of apical and basal dendritic arbors for reconstructed Tbr2 (red shading) and non-Tbr2 (green shading) spiny stellate neurons indicated with dark and light shading, respectively. Scale bar, 50 μm. (C) Confocal image of a representative pyramidal neuron. Scale bar, 100 μm. (D) Two-dimensional projections of convex hull volumes for pyramidal neurons. Scale bar, 100 μm. (E) Number of dendritic nodes and ends and total dendritic length of the apical dendrites of spiny stellate; and (F) pyramidal neurons. (G) Sholl analysis showed greater dendritic complexity of the apical arbor in the non-Tbr2 compared to Tbr2 lineage spiny stellate neurons. (H) Sholl analysis showed increased complexity of the proximal apical dendritic arbor in the Tbr2 lineage pyramidal neurons. (I) Number of dendritic nodes and ends and total dendritic length of the basal dendrites of spiny stellate; and (J) pyramidal neurons. (K) Sholl analysis showed no difference in the dendritic complexity of the basal arbor between Tbr2 and non-Tbr2 spiny stellate; and (L) pyramidal neurons. Figure 4. View largeDownload slide Spiny stellate and pyramidal neurons have different structural properties depending on their lineage. (A) Confocal image of a representative spiny stellate neuron. Scale bar, 50 μm. (B) Two-dimensional projections of convex hull volumes of apical and basal dendritic arbors for reconstructed Tbr2 (red shading) and non-Tbr2 (green shading) spiny stellate neurons indicated with dark and light shading, respectively. Scale bar, 50 μm. (C) Confocal image of a representative pyramidal neuron. Scale bar, 100 μm. (D) Two-dimensional projections of convex hull volumes for pyramidal neurons. Scale bar, 100 μm. (E) Number of dendritic nodes and ends and total dendritic length of the apical dendrites of spiny stellate; and (F) pyramidal neurons. (G) Sholl analysis showed greater dendritic complexity of the apical arbor in the non-Tbr2 compared to Tbr2 lineage spiny stellate neurons. (H) Sholl analysis showed increased complexity of the proximal apical dendritic arbor in the Tbr2 lineage pyramidal neurons. (I) Number of dendritic nodes and ends and total dendritic length of the basal dendrites of spiny stellate; and (J) pyramidal neurons. (K) Sholl analysis showed no difference in the dendritic complexity of the basal arbor between Tbr2 and non-Tbr2 spiny stellate; and (L) pyramidal neurons. Spiny Stellate Neurons We found that spiny stellate morphology differed depending on the lineage of the neuron. For example, Tbr2-derived spiny stellate neurons (n = 16) possessed less complex apical dendritic trees with fewer dendritic nodes and ends, and shorter total dendritic lengths compared to non-Tbr2-derived spiny stellate neurons (n = 13) (P < 0.05) (Fig. 4E,G). Furthermore, the convex hull (CH) volume of the apical dendritic arbor was significantly smaller for Tbr2-derived neurons (CH volume Tbr2 = 92 710 ± 24 069 μm3, non-Tbr2 = 201 796 ± 33 009 μm3, P < 0.01). In contrast, no differences were found in the basal dendrites between lineages (Fig. 4I,K; Fig. S2). Layer 4 Pyramidal Neurons In contrast to spiny stellate neurons, pyramidal neurons from the Tbr2 lineage (n = 13) and non-Tbr2 lineage (n = 9) did not differ with regard to their total mean apical dendritic complexity or length (Fig. 4F). However, detailed Sholl analyses revealed that Tbr2 pyramidal neurons had a higher degree of complexity (number of intersections) in the proximal apical dendritic branches (Fig. 4H). Specifically, the segment of the apical arbor located within the most proximal 100 microns—located within layer 4 specifically—was more complex in the Tbr2 group (Fig. 4H). As with spiny stellate neurons, there were no statistically significant differences in the mean length, complexity (Fig. 4J,L) or CH volume of the basal arbor between lineages (Fig. S2). Differential Polarization of the Dendritic Arbors of Tbr2 and Non-Tbr2 Spiny Stellate Neurons Previous studies have estimated that approximately 30% of spiny stellate neurons in layer 4 of the barrel cortex are polarized (Staiger et al. 2004) and direct their dendrites towards thalamic input within the barrel core (Valverde 1968; Borges and Berry 1976; Harris and Woolsey 1981; Egger et al. 2008). The asymmetry in the dendritic arbors of these neurons is thought to be due to a specific interplay between afferent input and dendritic topology resulting from selective dendritic pruning and outgrowth (Greenough and Chang 1988). We assessed the degree of polarization of spiny stellate neurons using vector analyses of dendritic length and direction (Fig. 5). The degree of horizontal polarity was significantly higher in Tbr2 versus non-Tbr2 spiny stellate neurons (P < 0.01) (Fig. 5C). Tbr2-derived neurons also had a greater percentage of total dendritic length inside the barrel than non-Tbr2 lineage neurons (Tbr2 lineage = 88.85 ± 3.27%, non-Tbr2 lineage = 72.91 ± 4.87%, P < 0.05). Since polarization is directed specifically towards the VGLUT2+ barrel core, these results are consistent with the idea that Tbr2 lineage spiny stellate neurons undergo enhanced dendritic remodeling to specialize in the reception of thalamic inputs. Figure 5. View largeDownload slide Tbr2 lineage spiny stellate neurons are more polarized towards the barrel core than are non-Tbr2 lineage neurons. (A) Examples of dendritic reconstructions of non-Tbr2 (left) and Tbr2 (right) derived spiny stellate neurons. Scale bar, 200 μm. (B) Polar histograms of dendritic length of neurons in A. (C) Box and Whisker and vertical scatter plots of the “X” (left) and “Y” (right) projections of the summation vectors of the polar histograms; *P < 0.05, **P < 0.01. Figure 5. View largeDownload slide Tbr2 lineage spiny stellate neurons are more polarized towards the barrel core than are non-Tbr2 lineage neurons. (A) Examples of dendritic reconstructions of non-Tbr2 (left) and Tbr2 (right) derived spiny stellate neurons. Scale bar, 200 μm. (B) Polar histograms of dendritic length of neurons in A. (C) Box and Whisker and vertical scatter plots of the “X” (left) and “Y” (right) projections of the summation vectors of the polar histograms; *P < 0.05, **P < 0.01. Dendritic Spine Distributions and VGLUT2+ Inputs to Tbr2 and Non-Tbr2-derived Neurons Overall, our analyses of dendritic topology revealed a number of lineage-specific differences in the structure of both layer 4 spiny stellate and pyramidal neurons. The morphology of neurons in layer 4 is influenced by excitatory synaptic inputs to dendritic spines and especially by thalamocortical inputs (Vercelli et al. 1992; Callaway and Borrell 2011). Therefore, we assessed the dendritic spines (Fig. 6) and putative thalamic inputs (Fig. 7) to Tbr2 and non-Tbr2-derived spiny stellate and pyramidal neurons. Data on the properties of all Tbr2 and non-Tbr2 neurons independent of morphological type are provided in Supplementary Table S1. Figure 6. View largeDownload slide Tbr2 and non-Tbr2 lineage pyramidal but not stellate neurons differ in apical spine density. (A) High-resolution confocal image of a dendritic segment of a filled neuron. Spine subtypes indicated with colored asterisks, with thin = green; filopodia = yellow; mushroom = red; stubby = blue. Scale bar, 5 μm. (B) Diagram showing criteria used to classify spine subtypes. (C) Mean total spine density and spine density Sholl analyses for the apical dendrites of spiny stellate; and (D) pyramidal neurons; *P < 0.05. (E) Mean spine density by subtype in stellate; and (F) pyramidal neurons; *P < 0.05. (G) Examples of reconstructions of pyramidal neurons showing the portion of the apical dendrite that extends to layers 2–3. (H) Sholl analysis of spine density in the portion of the apical dendritic arbor located in layers 2–3. Figure 6. View largeDownload slide Tbr2 and non-Tbr2 lineage pyramidal but not stellate neurons differ in apical spine density. (A) High-resolution confocal image of a dendritic segment of a filled neuron. Spine subtypes indicated with colored asterisks, with thin = green; filopodia = yellow; mushroom = red; stubby = blue. Scale bar, 5 μm. (B) Diagram showing criteria used to classify spine subtypes. (C) Mean total spine density and spine density Sholl analyses for the apical dendrites of spiny stellate; and (D) pyramidal neurons; *P < 0.05. (E) Mean spine density by subtype in stellate; and (F) pyramidal neurons; *P < 0.05. (G) Examples of reconstructions of pyramidal neurons showing the portion of the apical dendrite that extends to layers 2–3. (H) Sholl analysis of spine density in the portion of the apical dendritic arbor located in layers 2–3. Figure 7. View largeDownload slide VGLUT2+ appositions onto dendritic spines of Tbr2 and non-Tbr2 pyramidal neurons. (A) Image of a filled neuron (green) located in the barrel field as shown by VGLUT2 immunoreactivity (red). Cortical Layers: L1, L2–3, L4, L5. Scale bar, 100 μm. (B) High-resolution confocal micrograph of a filled neuron and VGLUT2+ boutons. Scale bar, 500 μm. (C) Dendritic segment of the neuron in B (segment area marked by dashed box in B), asterisks mark examples of appositions. Scale bar, 5 μm. (D) Sequential optical sections through a VGLUT2+ apposition, top to bottom. Scale bar, 2 μm. (E,F) Total VGLUT2/spine ratio (left) and Sholl analysis of VGLUT2/spine ratio (right) of the apical dendrite (E) and basal dendrite (F) of pyramidal neurons; *P < 0.05. (G) Representative reconstructions of Tbr2 and non-Tbr2-derived pyramidal neurons showing the apical dendrite in red and green, respectively, and their location relative to the barrel field. (H,J) Total spine estimate (left) and total mushroom spine estimate (right) of the portion of the apical dendrite that locates in layers 2–3 (H) and layer 4 (J). (I,K) Estimate of total appositions (left) and appositions on mushroom spines (right) of the portion of the apical dendrite that locates in layers 2–3 (I) and layer 4 (K); (*P < 0.05, **P < 0.01, ***P < 0.001). Figure 7. View largeDownload slide VGLUT2+ appositions onto dendritic spines of Tbr2 and non-Tbr2 pyramidal neurons. (A) Image of a filled neuron (green) located in the barrel field as shown by VGLUT2 immunoreactivity (red). Cortical Layers: L1, L2–3, L4, L5. Scale bar, 100 μm. (B) High-resolution confocal micrograph of a filled neuron and VGLUT2+ boutons. Scale bar, 500 μm. (C) Dendritic segment of the neuron in B (segment area marked by dashed box in B), asterisks mark examples of appositions. Scale bar, 5 μm. (D) Sequential optical sections through a VGLUT2+ apposition, top to bottom. Scale bar, 2 μm. (E,F) Total VGLUT2/spine ratio (left) and Sholl analysis of VGLUT2/spine ratio (right) of the apical dendrite (E) and basal dendrite (F) of pyramidal neurons; *P < 0.05. (G) Representative reconstructions of Tbr2 and non-Tbr2-derived pyramidal neurons showing the apical dendrite in red and green, respectively, and their location relative to the barrel field. (H,J) Total spine estimate (left) and total mushroom spine estimate (right) of the portion of the apical dendrite that locates in layers 2–3 (H) and layer 4 (J). (I,K) Estimate of total appositions (left) and appositions on mushroom spines (right) of the portion of the apical dendrite that locates in layers 2–3 (I) and layer 4 (K); (*P < 0.05, **P < 0.01, ***P < 0.001). Mean spine number, density, distribution and subtype proportions did not differ between basal dendrites of Tbr2 and non-Tbr2-derived spiny stellate neurons, and this was also the case for lineage-specific pyramidal neurons (Fig. S2). Furthermore, these parameters did not differ for the apical dendrites of spiny stellate neurons from the 2 lineages (Fig. 6C,E). By marked contrast, the apical dendrites of non-Tbr2-derived pyramidal neurons possessed a significantly higher density of spines compared to those of the Tbr2 lineage (P < 0.05) (Fig. 6D, left panel). Sholl analysis revealed that this difference was due to an increased density of spines in the mid-apical dendritic region (Fig. 6D) and to the thin spine subtype specifically (Fig. 6F). Because the apical dendrites of pyramidal neurons extend outside of layer 4, we quantified the spine density and distribution on apical dendrites located within layers 2–3 (Fig. 6G,H). This analysis revealed that the higher density of apical dendritic spines of non-Tbr2 pyramidal neurons is restricted to the segments located in layers 2–3 (Fig. 6H). Assessment of the density and location of VGLUT2+ thalamic appositions revealed no significant differences in apposition density on spines (in total, or on any morphological subtype), between lineages in either the pyramidal or stellate neuron groups (Fig. 7; Fig. S3). To obtain a general estimate of the proportion of excitatory input that was thalamocortical, we calculated the ratio of spines with VGLUT2+ appositions to total spines on a given dendrite. Tbr2-derived pyramidal neurons had a higher ratio of VGLUT2+ spine/total spine appositions in the proximal apical arbor, than did their non-Tbr2 counterparts (Fig. 7E). By contrast, the apical arbor of spiny stellates had the same VGLUT2+ spine/total spine ratio in both Tbr2 and non-Tbr2 lineages (Fig. S4), as did the basal arbors in both the pyramidal (Fig. 7F) and spiny stellate neurons (Fig. S4) from both lineages. The proximal apical dendritic compartment of pyramidal neurons from the Tbr2 lineage thus exhibited both higher VGLUT2+ apposition proportion (Fig. 7E) and higher branching complexity (Fig. 4H) compared to those from the non-Tbr2 lineage. We obtained a general estimate of the total number of spines and VGLUT2+ appositions across the entire apical arbor for each cell by extrapolating the values for spine and apposition density (total and by subtype) from individual dendrites to the total dendritic length within layer 4 (Fig. 7G–K). While the total spine number in the layer 4 portion of the pyramidal apical dendrite (Fig. 7G) did not differ between lineages (Fig. 7J, left; P = 0.31), the number of mushroom spines was 2.5 times higher in the Tbr2 compared to non-Tbr2 lineage (P < 0.001) (Fig. 7J, right). This is potentially functionally relevant since among spine subtypes, mushroom spines contain the highest number of glutamate receptors and are the site of potent synaptic excitation and plasticity (Hering and Sheng 2001; Rollenhagen and Lubke 2006; Bourne and Harris 2008). Furthermore, the number of VGLUT2+ boutons apposed to mushroom spines was also significantly higher in the apical dendrites of Tbr2-derived pyramidal neurons (Fig. 7K; P < 0.01). Higher Action Potential Firing Rates in both Spiny Stellate and Pyramidal Cells of the Tbr2 Compared to Non-Tbr2 Lineage Whole-cell patch clamp recordings of lineage-identified neurons were employed to compare the electrophysiological properties of Tbr2 and non-Tbr2 spiny stellate and pyramidal neurons. Passive membrane (Fig. 8A,B) and excitatory synaptic response (Fig. S6) properties of Tbr2 versus non-Tbr2 neurons did not differ in either the spiny stellate or pyramidal neuron groups (see also Supplementary Table S2). However, rheobase was significantly lower in Tbr2 compared to non-Tbr2 pyramidal (but not spiny stellate) neurons (Fig. 8B). Figure 8. View largeDownload slide Neurons from Tbr2 and non-Tbr2 lineages in layer 4 exhibit distinct action potential firing properties. (A) Top panel—representative spiny stellate neurons of the non-Tbr2 and Tbr2 lineages from which recordings were obtained. Bottom panel—box plots of resting membrane potential (left), input resistance (middle) and rheobase (right) of spiny stellate neurons of the 2 lineages. Scale bar, 100 μm. (B) Top panel—representative pyramidal neurons of the non-Tbr2 and Tbr2 lineages from which recordings were obtained. Bottom panel—box plots of resting membrane potential (left), input resistance (middle) and rheobase (right) of pyramidal neurons of the 2 lineages. Scale bar, 100 μm. (C) Left panel—representative voltage responses of spiny stellate neurons of the 2 lineages to a +175 pA current step. Scale bar, 20 mV/500 ms. Right panel—plot of the mean number of APs evoked by increasing current steps for stellate cells. (D) Left panel—representative voltage responses by pyramidal neurons of the 2 lineages to a +175 pA current step. Scale bar, 20 mV/500 ms. Right panel—plot of the mean number of APs evoked by increasing current steps for pyramidal cells. (E) Plots of the number of APs evoked by increasing current steps for all Tbr2 and non-Tbr2 neurons, with morphological types pooled for each lineage. Top panel shows data from individual cells, bottom panel shows mean data. (F) Box plot of the difference between time to last spike and time to first spike for all Tbr2 and all non-Tbr2 neurons; *P < 0.05. (G) Plots of number of evoked APs and time during the 2 s current pulse at which they occurred. Top panel shows data from individual cells, bottom panel shows mean data; *P < 0.05. Figure 8. View largeDownload slide Neurons from Tbr2 and non-Tbr2 lineages in layer 4 exhibit distinct action potential firing properties. (A) Top panel—representative spiny stellate neurons of the non-Tbr2 and Tbr2 lineages from which recordings were obtained. Bottom panel—box plots of resting membrane potential (left), input resistance (middle) and rheobase (right) of spiny stellate neurons of the 2 lineages. Scale bar, 100 μm. (B) Top panel—representative pyramidal neurons of the non-Tbr2 and Tbr2 lineages from which recordings were obtained. Bottom panel—box plots of resting membrane potential (left), input resistance (middle) and rheobase (right) of pyramidal neurons of the 2 lineages. Scale bar, 100 μm. (C) Left panel—representative voltage responses of spiny stellate neurons of the 2 lineages to a +175 pA current step. Scale bar, 20 mV/500 ms. Right panel—plot of the mean number of APs evoked by increasing current steps for stellate cells. (D) Left panel—representative voltage responses by pyramidal neurons of the 2 lineages to a +175 pA current step. Scale bar, 20 mV/500 ms. Right panel—plot of the mean number of APs evoked by increasing current steps for pyramidal cells. (E) Plots of the number of APs evoked by increasing current steps for all Tbr2 and non-Tbr2 neurons, with morphological types pooled for each lineage. Top panel shows data from individual cells, bottom panel shows mean data. (F) Box plot of the difference between time to last spike and time to first spike for all Tbr2 and all non-Tbr2 neurons; *P < 0.05. (G) Plots of number of evoked APs and time during the 2 s current pulse at which they occurred. Top panel shows data from individual cells, bottom panel shows mean data; *P < 0.05. In contrast to the similarities in passive membrane and synaptic properties, neurons showed marked lineage-specific differences in repetitive action potential firing properties regardless of morphological type (Fig. 8C–G). Specifically, in both spiny stellate and pyramidal neurons, the non-Tbr2 lineage group showed a high degree of adaptation of action potential firing, while Tbr2 lineage group did not (Fig. 8C–G). Furthermore, the Tbr2 lineage neurons fired a significantly higher number of action potentials during 2 s current steps at every current level (Fig. 8C–E; P < 0.05). The difference in number of action potentials across the 2 s steps was due to a higher degree of action potential firing rate adaptation in non-Tbr2 compared to Tbr2-derived neurons, since instantaneous firing frequency (first 20 ms of the step) did not differ (Supplementary Table S3). Multiple comparisons (across groups by lineage and morphological type, Figs. 8C,D and S5) and comparisons of all Tbr2 and non-Tbr2 neurons independent of morphological type (Fig. 8E) revealed that these differences in repetitive firing properties were due to lineage and not to morphological type. Conversely, neurons of distinct morphological class from the same lineage (i.e., Tbr2 pyramidal vs. Tbr2 stellate neurons) did not differ with regard to these repetitive firing properties (Fig. S5). To quantitate the differences in firing rate adaptation between Tbr2 and non-Tbr2 lineage neurons, we plotted the difference between the time to first and last spikes (Fig. 8F) and assessed both early and late firing frequency adaptation values (Supplementary Table S3). The interval between the first spike and last spike during a 2 s current pulse delivered at rheobase was significantly shorter in non-Tbr2 neurons (Fig. 8F), which typically ceased firing entirely mid-way through the current step (Fig. 8F,G) and could not be driven to fire with increasing current steps (beyond 225 pA). The marked adaptation in non-Tbr2 neurons, that occurred regardless of morphological type, is further illustrated by the individual and mean data for time of spike occurrence versus each spike occurring across the 2 s train (Fig. 8G). Discussion It is now widely accepted that both intrinsic and extrinsic cues are necessary for the development of cortical layers and areas. Graded transcription factor expression within dividing precursors and activity dependent cues provided by ingrowing afferents collectively generate the maps of functional cortical domains. However, the precise nature of these cues and how they combine to yield cortical complexity has remained elusive. Several studies have suggested a model of temporal fate restriction whereby dividing RGCs change molecularly over time to produce neurons that localize to different layers of the neocortex (Guo et al. 2013; Eckler et al. 2015; Desai and McConnell 2000; Shen et al. 2006). Other work has suggested that newborn excitatory neurons are fated to a specific lamina upon their terminal division (McConnell 1988; Alfano et al. 2014) and are thought to pass on positional (areal) information linking the germinal zone to the growing cortex since they migrate radially on the fibers of the RGC stem cells (Rakic 1972, 1988; O’Leary and Borngasser 2006). While these models potentially explain how different laminae are generated over time and how nascent cortical regions may be patterned, they do not account for how the different neuron types found within the local neighborhood of individual lamina are generated, nor how cellular heterogeneity is modified between cortical areas. Precursor Class Programs Neuronal Character During Cortical Plate Formation Here we demonstrate that key features of a temporal and spatial program of neocortical development are imparted by different contemporaneous precursor groups throughout neurogenesis. Using genetic fate-mapping, we show that neurons with distinct electrophysiological and morphological properties segregate by lineage. While this work confirms certain aspects of our previous findings in layers 2–3 of the frontal cortex (Tyler et al. 2015), the current analysis of layer 4, in a different region of cortex, highlights important lamina and area-specific precursor effects on neuronal heterogeneity. Together, these results show that different precursor groups do not merely amplify the output of parental RGCs, but instead generate neuronal diversity as each layer is produced. Our studies suggest that this process is modified in specific cortical areas, but the results elucidate several commonalities across different lamina. Two major cellular properties appear to be shaped by lineage, irrespective of which lamina is being produced: the morphology of the apical dendrite and action potential firing patterns. In particular, Tbr2 lineage pyramidal neurons in layers 2–3 and Tbr2 stellate neurons in layer 4 exhibit more simplified architecture of the apical dendrite than do non-Tbr2 lineage neurons. Tbr2 lineage neurons also fire action potentials at a higher rate (layers 2–3) or with lower degrees of adaptation (layer 4), regardless of their morphological subtype. We also show that these lineage-based activity and morphological programs segregate daughter cells in ways that may allow them to differentially interact with extrinsic cues provided by the multiple afferent inputs that impinge on the cortex. Interestingly, the properties of cells derived from the 2 lineages do not form 2 discrete populations of neurons, but rather, their properties mostly lie on different ends of a wide spectrum and can overlap in the middle of this range. This continuum of properties suggests that lineage of origin interacts with other factors to define neuron shape and function. It is also important to note that our fate-mapping technique does not fully demarcate all of the individual groups of neocortical progenitors, but rather segregates 2 major precursor pools to show that lineage identity deeply impacts the fate of neuronal progeny. Layer-specific Programming by Neural Precursors Our fate-mapping studies show that both Tbr2 and non-Tbr2 expressing precursors produce spiny stellate and pyramidal neurons in layer 4. This suggests that broad cues can act on all precursors at a specific developmental stage to yield appropriate classes of neurons for each layer. However, in addition to this basic developmental program, neurons derived from separate lineages differ in fundamental ways from one another even though they may be assigned to the same morphological subclass. For example, Tbr2 expressing precursors impart unique properties to their spiny stellate daughter cells compared to those from non-Tbr2 expressing precursors. As each cortical layer is produced, this process ultimately contributes to greater intralaminar neuronal diversity. The results indicate that while firing rate and apical dendritic complexity are the main parameters influenced by lineage identity, the nature of these parameters are modified in a layer-specific manner. Thus, while Tbr2 lineage neurons are more excitable than non-Tbr2 lineage neurons in both layers 2–3 and 4, the mechanisms by which they express this phenotype are different across layers. Specifically, higher firing rates observed for Tbr2 versus non-Tbr2 lineage neurons in layers 2–3 correlate with a higher input resistance. By contrast the higher firing rate of layer 4 Tbr2-derived neurons was not associated with input resistance but with decreased action potential adaptation compared to non-Tbr2 lineage neurons. This difference in firing frequency adaptation may result from differential activation of distinct potassium conductances in the 2 populations (Connor and Stevens 1971), a possibility that will be pursued in future studies. Similarly, we observed that lineage influenced the morphology of neurons, specifically with respect to topology of the apical dendrite. In layers 2–3, neurons derived from Tbr2 precursors had less arborized apical dendrites, while Tbr2-derived pyramidal neurons in layer 4 were more arborized proximal to the soma. Importantly, the intrinsic firing properties of a neuron dictate the dynamic range and fidelity with which it can transmit a signal, and changes in the morphometry of the apical dendrite determine the inputs that a given neuron can sample in the laminated cortex (Spruston 2008). Thus, our evidence strongly argues that neuronal precursor lineage significantly impacts the assembly of neocortical microcircuitry. Distinct Neural Precursor Groups Differentially Contribute to Barrel Topography and Connectivity Although several recent studies have addressed the contribution of IPCs to the laminar organization of the neocortex (Tarabykin et al. 2001; Kowalczyk et al. 2009; Vasistha et al. 2015; Mihalas et al. 2016), the present study is the first to assess the contribution of progenitor diversity to columnar organization. The readily apparent columnar segregation of thalamic inputs in the barrel cortex allowed us to do so. In contrast to layers 2–3, where pyramidal neurons from Tbr2 and non-Tbr2 lineages were intermingled and had a similar distribution across the layer, neurons derived from these lineages in layer 4 differed in their intralaminar localization, with Tbr2 lineage neurons residing deeper within the layer. This difference appears at the end of the first postnatal week, after the neurons have settled within the cortical plate. This suggests that the final positioning of neurons from each lineage within layer 4 is not due to differences in birth date or migration rates. The deeper settling pattern, combined with the lower rates of adaptation and differences in connectivity evident in Tbr2 lineage neurons, suggests that sublayers of processing may exist within layer 4 of the mouse barrel cortex, as has been shown in the primate visual cortex (Nassi and Callaway 2009). Furthermore, Tbr2-derived neurons preferentially reside in the barrel wall, when compared to their non-Tbr2-derived counterparts. This reveals that Tbr2 expressing precursors generate neurons with a specific role in the laminar and columnar arrangement of the neocortex, rather than generating progeny with a random distribution throughout the barrel field. In support of this observation, previous studies have shown that the barrel map is disrupted in Tbr2 knockout mice (Arnold et al. 2008; Elsen et al. 2013). Our results suggest that the absence of Tbr2-derived neurons would lead to a biased loss of cells in the barrel wall, thus altering the periphery-related patterning of the barrel. Further, the higher proportion of Tbr2-derived neurons in the barrel wall may have important consequences for how thalamic information is processed in layer 4, since the 2 barrel compartments receive inputs from distinct thalamic nuclei carrying different modalities of somatosensation (Yu et al. 2006; Feldmeyer 2012). Neurons derived from the 2 lineages differed in dendritic topology and distribution of inputs, which together are suggestive of differences in connectivity. First, Tbr2-derived spiny stellate neurons show a higher degree of horizontal polarization, which has been associated with a specialization in the reception of thalamocortical afferents (Valverde 1968; Borges and Berry 1976; Harris and Woolsey 1981; Egger et al. 2008). Tbr2-derived pyramidal neurons also exhibited profuse branching of the portion of the apical dendrite located within layer 4. Because thalamic afferents influence both the selective pruning process (Vercelli et al. 1992; Callaway and Borrell 2011) and the orientation of dendrites toward the barrel core (Valverde 1968; Borges and Berry 1976; Harris and Woolsey 1981; Egger et al. 2008), we determined the weight of VGLUT2-positive (primarily thalamic) appositions onto the spines of neurons from both lineages. The data are consistent with the idea that the apical dendrite of Tbr2-derived pyramidal neurons may receive greater thalamic input in layer 4 than that of non-Tbr2-derived pyramidal neurons. Thus, lineage identity might influence the pattern of extrinsic connectivity impinging on the dendritic arbor. Future studies will explore the functional relevance of these anatomical differences in thalamic appositions. Although the total number of thalamic boutons apposed to the dendritic arbor was similar in both lineages, differences in the apical arbor may result in differential dendritic filtering of synaptic inputs. Alternatively, the differences described here may be reminiscent of a differential influence of TCA in the pruning of the apical dendrite during development. The IPC Program Model Based on our findings, we propose a new model to explain the origin of neuronal heterogeneity within lamina and across cortical areas (Fig. 9). We show that IPCs play a large role in this process in a manner that appears to be independent of their apical mother cells. These data prompt an important modification to a model in which neocortical growth is thought to be primarily due to temporal changes in intrinsic programs of aRGCs. In the RGC fate restriction model, aRGCs are successively modified to produce neurons for each lamina over time. This model assumes that intermediate precursors merely boost neuronal production capacity and neglects any roles the individual classes of precursors derived from the aRGCs may play; in so doing, it assumes that intermediate precursors convey the same information provided by the aRGCs. Recent experiments indicate that Drosophila intermediate neural progenitors generate diverse neuron types through a process of sequential transcription factor expression (Bayraktar and Doe 2013). The current data demonstrate that mammalian IPCs also change temporally during neurogenesis and that their changes are distinct from those occurring within the aRGCs (Fig. 9B). These divergent intrinsic programs lead to production of a varied neuronal pool for each lamina. Figure 9. View largeDownload slide Proposed models of lineage-specified connectivity in layer 4 and IPC neuronal production. (A) Schematic summary of the major differences between Tbr2 and non-Tbr2-derived pyramidal (left) and spiny stellate (right) neurons in layer 4 of the mouse barrel cortex. Boundaries of layers (L1, L2–3, L4, L5) and barrel core compartments (light blue areas) are delineated by dashed lines. At P21, Tbr2 lineage neurons are situated deeper in layer 4 compared to non-Tbr2 lineage neurons. The proximal apical dendrites of Tbr2 pyramidal neurons are more arborized in layer 4 and receive a higher number of putative inputs from VGLUT2+ thalamocortical afferents (TCA, blue dots) on mushroom spines (large spines) compared to non-Tbr2 pyramidal neurons (a1). The apical dendrites of non-Tbr2 pyramidal neurons have a higher density of thin spines (small spines) in layers 2-3 (a2). Tbr2 spiny stellate neurons have less complex apical dendrites and their arbors are more polarized towards the barrel core (a3). Lineage-specific differences in action potential firing properties independent of morphological type were found. For both morphological types, Tbr2 lineage neurons fired a higher number of action potential responses and exhibited lower firing rate adaptation than non-Tbr2 neurons (a4). (B) Schematic of our proposed model of the role of IPCs in establishing neuronal heterogeneity within and across cortical laminae during neurogenesis. Layers are labeled (L1, L2–3, L4, L5, L6) as they are populated by daughter neurons at each time point. Our data demonstrate that simultaneously generated neurons have distinct properties depending on their precursor of origin, and that neurons from the same lineage are different depending on their birth date. Figure 9. View largeDownload slide Proposed models of lineage-specified connectivity in layer 4 and IPC neuronal production. (A) Schematic summary of the major differences between Tbr2 and non-Tbr2-derived pyramidal (left) and spiny stellate (right) neurons in layer 4 of the mouse barrel cortex. Boundaries of layers (L1, L2–3, L4, L5) and barrel core compartments (light blue areas) are delineated by dashed lines. At P21, Tbr2 lineage neurons are situated deeper in layer 4 compared to non-Tbr2 lineage neurons. The proximal apical dendrites of Tbr2 pyramidal neurons are more arborized in layer 4 and receive a higher number of putative inputs from VGLUT2+ thalamocortical afferents (TCA, blue dots) on mushroom spines (large spines) compared to non-Tbr2 pyramidal neurons (a1). The apical dendrites of non-Tbr2 pyramidal neurons have a higher density of thin spines (small spines) in layers 2-3 (a2). Tbr2 spiny stellate neurons have less complex apical dendrites and their arbors are more polarized towards the barrel core (a3). Lineage-specific differences in action potential firing properties independent of morphological type were found. For both morphological types, Tbr2 lineage neurons fired a higher number of action potential responses and exhibited lower firing rate adaptation than non-Tbr2 neurons (a4). (B) Schematic of our proposed model of the role of IPCs in establishing neuronal heterogeneity within and across cortical laminae during neurogenesis. Layers are labeled (L1, L2–3, L4, L5, L6) as they are populated by daughter neurons at each time point. Our data demonstrate that simultaneously generated neurons have distinct properties depending on their precursor of origin, and that neurons from the same lineage are different depending on their birth date. Conclusion In sum, our findings demonstrate that neural precursor lineages exert a complex effect on the generation of cortical diversity that goes beyond mass production of neurons. The lineage effects on key cellular properties are area- and layer-specific and thus subject to temporal changes during embryonic development. In terms of the relationship between intrinsic and extrinsic cues during neocortical growth, the results outline how lineage identity plays a major role in providing a framework upon which the afferents carrying extrinsic cues differentially impinge. Funding This work was supported by grants from the National Institutes of Health NS051582 to T.F.H. and W.A.T., and NS089340 to T.F.H., W.A.T., and J.I.L.; Fulbright Commission Spain and Rafael del Pino Foundation to T.G-V. Notes We thank Dr. Howard Cabral for statistical advice, members of the T.F.H and J.I.L. laboratories for constructive discussions and Denston Carey Jr. and Afroze Shaikh for help with puncta and neuron counts. Conflict of interest: None declared. References Alfano C , Magrinelli E , Harb K , Hevner RF , Studer M . 2014 . Postmitotic control of sensory area specification during neocortical development . Nat Commun . 5 : 5632 . Google Scholar Crossref Search ADS PubMed Amatrudo JM , Weaver CM , Crimins JL , Hof PR , Rosene DL , Luebke JI . 2012 . Influence of highly distinctive structural properties on the excitability of pyramidal neurons in monkey visual and prefrontal cortices . J Neurosci . 32 ( 40 ): 13644 – 13660 . Google Scholar Crossref Search ADS PubMed Arnold SJ , Huang GJ , Cheung AF , Era T , Nishikawa S , Bikoff EK , Molnar Z , Robertson EJ , Groszer M . 2008 . The T-box transcription factor Eomes/Tbr2 regulates neurogenesis in the cortical subventricular zone . Genes Dev . 22 : 2479 – 2484 . Google Scholar Crossref Search ADS PubMed Bayraktar OA , Doe CQ . 2013 . Combinatorial temporal patterning in progenitors expands neural diversity . Nature . 498 : 449 – 455 . Google Scholar Crossref Search ADS PubMed Betizeau M , Cortay V , Patti D , Pfister S , Gautier E , Bellemin-Menard A , Afanassieff M , Huissoud C , Douglas RJ , Kennedy H , et al. 2013 . Precursor diversity and complexity of lineage relationships in the outer subventricular zone of the primate . Neuron . 80 : 442 – 457 . Google Scholar Crossref Search ADS PubMed Borges S , Berry M . 1976 . Preferential orientation of stellate cell dendrites in the visual cortex of the dark-reared rat . Brain Res . 112 : 141 – 147 . Google Scholar Crossref Search ADS PubMed Bourne JN , Harris KM . 2008 . Balancing structure and function at hippocampal dendritic spines . Annu Rev Neurosci . 31 : 47 – 67 . Google Scholar Crossref Search ADS PubMed Callaway EM , Borrell V . 2011 . Developmental sculpting of dendritic morphology of layer 4 neurons in visual cortex: influence of retinal input . J Neurosci . 31 : 7456 – 7470 . Google Scholar Crossref Search ADS PubMed Connor JA , Stevens CF . 1971 . Prediction of repetitive firing behaviour from voltage clamp data on an isolated neurone soma . J Physiol . 213 : 31 – 53 . Google Scholar Crossref Search ADS PubMed Crimins JL , Rocher AB , Luebke JI . 2012 . Electrophysiological changes precede morphological changes to frontal cortical pyramidal neurons in the rTg4510 mouse model of progressive tauopathy . Acta Neuropathol . 122 ( 5 ): 551 – 564 . Google Scholar Crossref Search ADS Dehay C , Kennedy H . 2007 . Cell-cycle control and cortical development . Nat Rev Neurosci . 8 : 438 – 450 . Google Scholar Crossref Search ADS PubMed Desai AR , Mcconnell SK . 2000 . Progressive restriction in fate potential by neural progenitors during cerebral cortical development . Development . 127 : 2863 – 2872 . Google Scholar PubMed Eckler MJ , Nguyen TD , Mckenna WL , Fastow BL , Guo C , Rubenstein JL , Chen B . 2015 . Cux2-positive radial glial cells generate diverse subtypes of neocortical projection neurons and macroglia . Neuron . 86 : 1100 – 1108 . Google Scholar Crossref Search ADS PubMed Egger V , Nevian T , Bruno RM . 2008 . Subcolumnar dendritic and axonal organization of spiny stellate and star pyramid neurons within a barrel in rat somatosensory cortex . Cereb Cortex . 18 : 876 – 889 . Google Scholar Crossref Search ADS PubMed Elsen GE , Hodge RD , Bedogni F , Daza RA , Nelson BR , Shiba N , Reiner SL , Hevner RF . 2013 . The protomap is propagated to cortical plate neurons through an Eomes-dependent intermediate map . Proc Natl Acad Sci USA . 110 : 4081 – 4086 . Google Scholar Crossref Search ADS PubMed Englund C , Fink A , Lau C , Pham D , Daza RA , Bulfone A , Kowalczyk T , Hevner RF . 2005 . Pax6, Tbr2, and Tbr1 are expressed sequentially by radial glia, intermediate progenitor cells, and postmitotic neurons in developing neocortex . J Neurosci . 25 : 247 – 251 . Google Scholar Crossref Search ADS PubMed Feldmeyer D . 2012 . Excitatory neuronal connectivity in the barrel cortex . Front Neuroanat . 6 : 24 . Google Scholar Crossref Search ADS PubMed Feldmeyer D , Egger V , Lubke J , Sakmann B . 1999 . Reliable synaptic connections between pairs of excitatory layer 4 neurones within a single ‘barrel’ of developing rat somatosensory cortex . J Physiol . 521 ( Pt 1 ): 169 – 190 . Google Scholar Crossref Search ADS PubMed Fietz SA , Kelava I , Vogt J , Wilsch-Brauninger M , Stenzel D , Fish JL , Corbeil D , Riehn A , Distler W , Nitsch R , et al. 2010 . OSVZ progenitors of human and ferret neocortex are epithelial-like and expand by integrin signaling . Nat Neurosci . 13 : 690 – 699 . Google Scholar Crossref Search ADS PubMed Gal JS , Morozov YM , Ayoub AE , Chatterjee M , Rakic P , Haydar TF . 2006 . Molecular and morphological heterogeneity of neural precursors in the mouse neocortical proliferative zones . J Neurosci . 26 : 1045 – 1056 . Google Scholar Crossref Search ADS PubMed Greenough WT , Chang FL . 1988 . Dendritic pattern formation involves both oriented regression and oriented growth in the barrels of mouse somatosensory cortex . Brain Res . 471 : 148 – 152 . Google Scholar Crossref Search ADS PubMed Guo C , Eckler MJ , Mckenna WL , Mckinsey GL , Rubenstein JL , Chen B . 2013 . Fezf2 expression identifies a multipotent progenitor for neocortical projection neurons, astrocytes, and oligodendrocytes . Neuron . 80 : 1167 – 1174 . Google Scholar Crossref Search ADS PubMed Gundersen HJ , Jensen EB . 1987 . The efficiency of systematic sampling in stereology and its prediction . J Microsc . 147 ( Pt 3 ): 229 – 263 . Google Scholar Crossref Search ADS PubMed Hansen DV , Lui JH , Parker PR , Kriegstein AR . 2010 . Neurogenic radial glia in the outer subventricular zone of human neocortex . Nature . 464 : 554 – 561 . Google Scholar Crossref Search ADS PubMed Harris RM , Woolsey TA . 1981 . Dendritic plasticity in mouse barrel cortex following postnatal vibrissa follicle damage . J Comp Neurol . 196 : 357 – 376 . Google Scholar Crossref Search ADS PubMed Hering H , Sheng M . 2001 . Dendritic spines: structure, dynamics and regulation . Nat Rev Neurosci . 2 : 880 – 888 . Google Scholar Crossref Search ADS PubMed Hevner RF , Miyashita-Lin E , Rubenstein JL . 2002 . Cortical and thalamic axon pathfinding defects in Tbr1, Gbx2, and Pax6 mutant mice: evidence that cortical and thalamic axons interact and guide each other . J Comp Neurol . 447 : 8 – 17 . Google Scholar Crossref Search ADS PubMed Johnson MB , Wang PP , Atabay KD , Murphy EA , Doan RN , Hecht JL , Walsh CA . 2015 . Single-cell analysis reveals transcriptional heterogeneity of neural progenitors in human cortex . Nat Neurosci . 18 : 637 – 646 . Google Scholar Crossref Search ADS PubMed Jones EG . 1975 . Varieties and distribution of non-pyramidal cells in the somatic sensory cortex of the squirrel monkey . J Comp Neurol . 160 : 205 – 267 . Google Scholar Crossref Search ADS PubMed Kawaguchi A , Ikawa T , Kasukawa T , Ueda HR , Kurimoto K , Saitou M , Matsuzaki F . 2008 . Single-cell gene profiling defines differential progenitor subclasses in mammalian neurogenesis . Development . 135 : 3113 – 3124 . Google Scholar Crossref Search ADS PubMed Kowalczyk T , Pontious A , Englund C , Daza RA , Bedogni F , Hodge R , Attardo A , Bell C , Huttner WB , Hevner RF . 2009 . Intermediate neuronal progenitors (basal progenitors) produce pyramidal-projection neurons for all layers of cerebral cortex . Cereb Cortex . 19 : 2439 – 2450 . Google Scholar Crossref Search ADS PubMed Li H , Crair MC . 2011 . How do barrels form in somatosensory cortex? Ann N Y Acad Sci . 1225 : 119 – 129 . Google Scholar Crossref Search ADS PubMed Lopez-Bendito G , Molnar Z . 2003 . Thalamocortical development: how are we going to get there? Nat Rev Neurosci . 4 : 276 – 289 . Google Scholar Crossref Search ADS PubMed Lund JS . 1984 . Spiny Stellate Neurons. In: Peters A , Jones EG , editors . Cerebral cortex : cellular components of the cerebral cortex, Vol. 1. New York : Plenum Press , p. 255 – 308 . Malatesta P , Hartfuss E , Gotz M . 2000 . Isolation of radial glial cells by fluorescent-activated cell sorting reveals a neuronal lineage . Development . 127 : 5253 – 5263 . Google Scholar PubMed McConnell SK . 1988 . Fates of visual cortical neurons in the ferret after isochronic and heterochronic transplantation . J Neurosci . 8 : 945 – 974 . Google Scholar Crossref Search ADS PubMed Mihalas AB , Elsen GE , Bedogni F , Daza RA , Ramos-Laguna KA , Arnold SJ , Hevner RF . 2016 . Intermediate progenitor cohorts differentially generate cortical layers and require Tbr2 for timely acquisition of neuronal subtype identity . Cell Rep . 16 : 92 – 105 . Google Scholar Crossref Search ADS PubMed Mizutani K , Yoon K , Dang L , Tokunaga A , Gaiano N . 2007 . Differential Notch signalling distinguishes neural stem cells from intermediate progenitors . Nature . 449 : 351 – 355 . Google Scholar Crossref Search ADS PubMed Nassi JJ , Callaway EM . 2009 . Parallel processing strategies of the primate visual system . Nat Rev Neurosci . 10 : 360 – 372 . Google Scholar Crossref Search ADS PubMed Nelson BR , Hodge RD , Bedogni F , Hevner RF . 2013 . Dynamic interactions between intermediate neurogenic progenitors and radial glia in embryonic mouse neocortex: potential role in Dll1-Notch signaling . J Neurosci . 33 : 9122 – 9139 . Google Scholar Crossref Search ADS PubMed Noctor SC , Flint AC , Weissman TA , Dammerman RS , Kriegstein AR . 2001 . Neurons derived from radial glial cells establish radial units in neocortex . Nature . 409 : 714 – 720 . Google Scholar Crossref Search ADS PubMed Noctor SC , Flint AC , Weissman TA , Wong WS , Clinton BK , Kriegstein AR . 2002 . Dividing precursor cells of the embryonic cortical ventricular zone have morphological and molecular characteristics of radial glia . J Neurosci . 22 : 3161 – 3173 . Google Scholar Crossref Search ADS PubMed Oberlaender M , De Kock CP , Bruno RM , Ramirez A , Meyer HS , Dercksen VJ , Helmstaedter M , Sakmann B . 2012 . Cell type-specific three-dimensional structure of thalamocortical circuits in a column of rat vibrissal cortex . Cereb Cortex . 22 : 2375 – 2391 . Google Scholar Crossref Search ADS PubMed O’Leary DD , Borngasser D . 2006 . Cortical ventricular zone progenitors and their progeny maintain spatial relationships and radial patterning during preplate development indicating an early protomap . Cereb Cortex . 16 ( Suppl 1 ): i46 – i56 . Google Scholar Crossref Search ADS PubMed O’Leary DD , Schlaggar BL , Tuttle R . 1994 . Specification of neocortical areas and thalamocortical connections . Annu Rev Neurosci . 17 : 419 – 439 . Google Scholar Crossref Search ADS PubMed Paxinos G , Franklin KBJ 2003 . The Mouse Brain in Stereotaxic Coordinates . Peterson DA . 1999 . Quantitative histology using confocal microscopy: implementation of unbiased stereology procedures . Methods . 18 : 493 – 507 . Google Scholar Crossref Search ADS PubMed Pollen AA , Nowakowski TJ , Shuga J , Wang X , Leyrat AA , Lui JH , Li N , Szpankowski L , Fowler B , Chen P , et al. 2014 . Low-coverage single-cell mRNA sequencing reveals cellular heterogeneity and activated signaling pathways in developing cerebral cortex . Nat Biotechnol . 32 : 1053 – 1058 . Google Scholar Crossref Search ADS PubMed Rakic P . 1972 . Mode of cell migration to the superficial layers of fetal monkey neocortex . J Comp Neurol . 145 : 61 – 83 . Google Scholar Crossref Search ADS PubMed Rakic P . 1988 . Specification of cerebral cortical areas . Science . 241 : 170 – 176 . Google Scholar Crossref Search ADS PubMed Rocher AB , Crimins JL , Amatrudo JM , Kinson MS , Todd-Brown MA , Lewis J , Luebke JI . 2010 . Structural and functional changes in tau mutant mice neurons are not linked to the presence of NFTs . Exp Neurol . 223 ( 2 ): 385 – 393 . Google Scholar Crossref Search ADS PubMed Rollenhagen A , Lubke JH . 2006 . The morphology of excitatory central synapses: from structure to function . Cell Tissue Res . 326 : 221 – 237 . Google Scholar Crossref Search ADS PubMed Schubert D , Kotter R , Zilles K , Luhmann HJ , Staiger JF . 2003 . Cell type-specific circuits of cortical layer IV spiny neurons . J Neurosci . 23 : 2961 – 2970 . Google Scholar Crossref Search ADS PubMed Shen Q , Wang Y , Dimos JT , Fasano CA , Phoenix TN , Lemischka IR , Ivanova NB , Stifani S , Morrisey EE , Temple S . 2006 . The timing of cortical neurogenesis is encoded within lineages of individual progenitor cells . Nat Neurosci . 9 : 743 – 751 . Google Scholar Crossref Search ADS PubMed Shitamukai A , Konno D , Matsuzaki F . 2011 . Oblique radial glial divisions in the developing mouse neocortex induce self-renewing progenitors outside the germinal zone that resemble primate outer subventricular zone progenitors . J Neurosci . 31 : 3683 – 3695 . Google Scholar Crossref Search ADS PubMed Spruston N . 2008 . Pyramidal neurons: dendritic structure and synaptic integration . Nat Rev Neurosci . 9 : 206 – 221 . Google Scholar Crossref Search ADS PubMed Staiger JF , Flagmeyer I , Schubert D , Zilles K , Kotter R , Luhmann HJ . 2004 . Functional diversity of layer IV spiny neurons in rat somatosensory cortex: quantitative morphology of electrophysiologically characterized and biocytin labeled cells . Cereb Cortex . 14 : 690 – 701 . Google Scholar Crossref Search ADS PubMed Stancik EK , Navarro-Quiroga I , Sellke R , Haydar TF . 2010 . Heterogeneity in ventricular zone neural precursors contributes to neuronal fate diversity in the postnatal neocortex . J Neurosci . 30 : 7028 – 7036 . Google Scholar Crossref Search ADS PubMed Sur M , Rubenstein JL . 2005 . Patterning and plasticity of the cerebral cortex . Science . 310 : 805 – 810 . Google Scholar Crossref Search ADS PubMed Tarabykin V , Stoykova A , Usman N , Gruss P . 2001 . Cortical upper layer neurons derive from the subventricular zone as indicated by Svet1 gene expression . Development . 128 : 1983 – 1993 . Google Scholar PubMed Tyler WA , Haydar TF . 2013 . Multiplex genetic fate mapping reveals a novel route of neocortical neurogenesis, which is altered in the Ts65Dn mouse model of Down syndrome . J Neurosci . 33 : 5106 – 5119 . Google Scholar Crossref Search ADS PubMed Tyler WA , Medalla M , Guillamon-Vivancos T , Luebke JI , Haydar TF . 2015 . Neural precursor lineages specify distinct neocortical pyramidal neuron types . J Neurosci . 35 : 6142 – 6152 . Google Scholar Crossref Search ADS PubMed Valverde F . 1968 . Structural changes in the area striata of the mouse after enucleation . Exp Brain Res . 5 : 274 – 292 . Google Scholar Crossref Search ADS PubMed Vasistha NA , Garcia-Moreno F , Arora S , Cheung AF , Arnold SJ , Robertson EJ , Molnar Z . 2015 . Cortical and clonal contribution of Tbr2 expressing progenitors in the developing mouse brain . Cereb Cortex . 25 : 3290 – 3302 . Google Scholar Crossref Search ADS PubMed Vercelli A , Assal F , Innocenti GM . 1992 . Emergence of callosally projecting neurons with stellate morphology in the visual cortex of the kitten . Exp Brain Res . 90 : 346 – 358 . Google Scholar Crossref Search ADS PubMed West MJ , Slomianka L , Gundersen HJ . 1991 . Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator . Anat Rec . 231 : 482 – 497 . Google Scholar Crossref Search ADS PubMed Woolsey TA , Van Der Loos H . 1970 . The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex. The description of a cortical field composed of discrete cytoarchitectonic units . Brain Res . 17 : 205 – 242 . Google Scholar Crossref Search ADS PubMed Yu C , Derdikman D , Haidarliu S , Ahissar E . 2006 . Parallel thalamic pathways for whisking and touch signals in the rat . PLoS Biol . 4 : e124 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Distinct Neocortical Progenitor Lineages Fine-tune Neuronal Diversity in a Layer-specific Manner
JF - Cerebral Cortex
DO - 10.1093/cercor/bhy019
DA - 2019-03-01
UR - https://www.deepdyve.com/lp/oxford-university-press/distinct-neocortical-progenitor-lineages-fine-tune-neuronal-diversity-7h2plNa5rQ
SP - 1121
VL - 29
IS - 3
DP - DeepDyve
ER -