1. Developmental Biology
  2. Neuroscience

Activity-dependent death of transient Cajal-Retzius neurons is required for functional cortical wiring

  1. Martina Riva
  2. Ioana Genescu
  3. Chloé Habermacher
  4. David Orduz
  5. Fanny Ledonne
  6. Filippo M Rijli
  7. Guillermina López-Bendito
  8. Eva Coppola
  9. Sonia Garel  Is a corresponding author
  10. Maria Cecilia Angulo  Is a corresponding author
  11. Alessandra Pierani  Is a corresponding author
  1. Institut Imagine, Université de Paris, France
  2. Institut Jacques Monod, CNRS UMR 7592, Université de Paris, France
  3. Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Université de Paris, France
  4. Institut de Biologie de l’École Normale Supérieure (IBENS), Département de Biologie, École Normale Supérieure, CNRS, INSERM, Université PSL, France
  5. INSERM U1128, France
  6. Friedrich Miescher Institute for Biomedical Research, Switzerland
  7. Universidad Miguel Hernandez, Spain
https://doi.org/10.7554/eLife.50503
Share this article
Cite this article
  1. Martina Riva
  2. Ioana Genescu
  3. Chloé Habermacher
  4. David Orduz
  5. Fanny Ledonne
  6. Filippo M Rijli
  7. Guillermina López-Bendito
  8. Eva Coppola
  9. Sonia Garel
  10. Maria Cecilia Angulo
  11. Alessandra Pierani
(2019)
Activity-dependent death of transient Cajal-Retzius neurons is required for functional cortical wiring
eLife 8:e50503.
https://doi.org/10.7554/eLife.50503
  2. Download BibTeX
  3. Download .RIS

Abstract

Programmed cell death and early activity contribute to the emergence of functional cortical circuits. While most neuronal populations are scaled-down by death, some subpopulations are entirely eliminated, raising the question of the importance of such demise for cortical wiring. Here, we addressed this issue by focusing on Cajal-Retzius neurons (CRs), key players in cortical development that are eliminated in postnatal mice in part via Bax-dependent apoptosis. Using Bax-conditional mutants and CR hyperpolarization, we show that the survival of electrically active subsets of CRs triggers an increase in both dendrite complexity and spine density of upper layer pyramidal neurons, leading to an excitation/inhibition imbalance. The survival of these CRs is induced by hyperpolarization, highlighting an interplay between early activity and neuronal elimination. Taken together, our study reveals a novel activity-dependent programmed cell death process required for the removal of transient immature neurons and the proper wiring of functional cortical circuits.

Introduction

An emerging player in the assembly of neuronal networks is programmed cell death (PCD). In the nervous system, programmed cell death (PCD) fine-tunes the density of neuronal populations by eliminating 20–40% of overproduced neurons (Fuchs and Steller, 2011; Causeret et al., 2018; Wong and Marín, 2019). Only few populations of the mouse cerebral cortex almost completely disappear during the first two postnatal weeks. Amongst these, Cajal-Retzius cells (CRs), the first-born cortical neurons lying in the superficial Layer I (LI), undergo extensive cell death in the mouse during the second postnatal week (Ledonne et al., 2016). The persistence of CRs during postnatal life is increased in malformations of cortical development (MCDs) and epilepsies thereby opening the intriguing possibility that the maintenance of CRs contributes to the dysfunction of cortical circuits (for review see Luhmann, 2013).

CR play pivotal roles at multiple steps of early cortical development, in addition to their best-known role in the control of radial migration (Ishii et al., 2016). They comprise three molecularly distinct subtypes which migrate from different sources that surround the cortical primordium: (i) septum and eminentia thalami-derived CRs of the ΔNp73/Dbx1 lineage (SE-CRs); (ii) hem-derived CRs of the ΔNp73/Wnt3a lineage (hem-CRs); (iii) pallial-subpallial boundary-derived CRs of the Dbx1 lineage (PSB-CRs) (Bielle et al., 2005; Yoshida et al., 2006; Tissir et al., 2009). Our previous work revealed that their specific embryonic distribution in distinct territories plays key functions in wiring of cortical circuits by controlling the size of functional areas, both primary and higher-order, as well as targeting of thalamocortical afferents (Griveau et al., 2010; Barber et al., 2015; Barber and Pierani, 2016). More recently, we showed that subtype-specific differences also exist in their elimination during early postnatal life, with SE- but not hem-derived CRs dying in a Bax-dependent manner, which is a critical player of the apoptotic pathway (Ledonne et al., 2016).

Before their disappearance, CRs express ionotropic glutamatergic and GABAergic receptors and are embedded into immature circuits where they mainly receive GABAergic synaptic inputs, suggesting that these transient cells might have an activity-dependent role in the development of cortical networks (Kirischuk et al., 2014). Consistently, we found that CRs density shapes axonal and dendritic outgrowth in LI and impacts onto the excitation/inhibition (E/I) ratio in upper cortical layers (de Frutos et al., 2016). Conversely, activity was also proposed as one of the drive of CR demise. Studies from Del Rı́o et al. (1996) first highlighted the role played by electrical activity on CR death in vitro. CRs were shown to display electrophysiological features of immature neurons (Kirischuk et al., 2014; Barber and Pierani, 2016), including the persistent depolarizing action of GABA which was suggested to depend on the maintenance of the chloride inward transporter NKCC1 and the absence of the outward transporter KCC2 (Mienville, 1998; Achilles et al., 2007; Pozas et al., 2008). Interestingly, pharmacological inhibition of activity and GABA signaling in vitro and global inactivation of NKCC1 in vivo were shown to reduce the death of CRs (Blanquie et al., 2017a; Blanquie et al., 2017b). However, very little is known on the role of electrical activity in CR subtype-specific death in vivo, as well as its contribution to the construction of functional and dysfunctional cortical circuits.

Here we show that hyperpolarization of CR neurons by Kir2.1-dependent expression prevented cell death of ΔNp73- but not Wnt3a-derived CRs, corresponding to SE-CRs. By comparing two different mutants in which CR death was similarly rescued, we found that abnormal SE-CR survival promotes exuberance of dendrites and spine density in pyramidal neurons in an activity-dependent manner. This results in an E/I imbalance due to an increase in the excitatory drive of upper cortical neurons. Our findings show that neuronal activity is involved in CR subtype-specific death and argue in favor of an unappreciated role of the disappearance of these transient neurons in controlling the morphology of pyramidal neurons and the functional properties of their cortical excitatory networks at postnatal stages.

Results

Subsets of CRs die in an activity-dependent manner

The ΔNp73cre/+ mouse line targets approximately 80% of CRs, namely hem-CRs (Wnt3a lineage) and SE-CRs (Tissir et al., 2009; Griveau et al., 2010; Yoshida et al., 2006). ΔNp73-CRs, but not Wnt3a-derived hem-CRs, were shown to be eliminated postnatally via Bax-dependent death, indicating that SE-CRs undergo apoptosis. However, the trigger of such apoptosis or the mechanisms regulating the death of other CR populations are still unknown (Ledonne et al., 2016). To decipher whether activity might regulate the death of specific subsets of CRs in vivo, we first overexpressed the hyperpolarizing channel Kir2.1 (R26Kir2.1mcherry/+) (Moreno-Juan et al., 2017) using the ΔNp73cre/+ line. Tracing of CRs at several postnatal stages in control and Kir2.1-expressing mice was performed using DsRed immunostainings to visualize tdTomato and mcherry reporters, respectively. We found that the density of CRs in the somatosensory barrel cortex was unchanged in these animals at postnatal day 7 (P7), that is before CRs undergo massive cell death (Figure 1A and B). In contrast, Kir2.1 channel overexpression in ΔNp73-CRs resulted in an increase of CRs with respect to controls in the somatosensory cortex at P15 and P25 (Figure 1A and B). Importantly, using whole-cell recording, we checked that rescued CRs were, as expected, hyperpolarized and displayed a decrease in the input resistance without exhibiting changes in action potential properties at both P15 and P25 (Figure 1—figure supplement 1A-C). Biocytin-filling and immunostaining further revealed that most morphological properties were preserved in Kir2.1-expressing CRs, apart from a reduced soma at P25 (Figure 1—figure supplement 1D). In particular, rescued CRs displayed similar branching length in LI (Figure 1—figure supplement 1B and E) and co-expressed Reelin (Reln) from P7 to P25 (Figure 1—figure supplement 1F and G). Taken together, these experiments show that hyperpolarization of ΔNp73-CRs does not drastically alter cardinal morphological features of CRs but prevents their complete elimination.

Figure 1 with 1 supplement see all
Download asset Open asset
Hyperpolarization induces the survival of SE-CRs.

(A) Confocal images of cortical sections from P7, P15 and P25 ΔNp73cre/+;R26mT/+ controls (left) and ΔNp73cre/+;R26Kir2.1/+ mutants (right) stained for DsRed (red) and Hoechst (blue). Arrowheads indicate DsRed+ CRs at P15 and P25. (B) Quantification of CR density (CRs/mm3) at the pial surface in the somatosensory (S1) cortex (P7: n = 6 for controls and n = 4 for mutants, p=0.716; P15: n = 4 for controls and n = 10 for mutants, p=0.001; P25: n = 3 for controls and n = 8 for mutants, p=0.006). (C) Confocal images of cortical sections from P7, P15 and P25 Wnt3acre/+;R26mT/+ controls (left) and Wnt3acre/+;R26Kir2.1/+ mutants (right) stained for DsRed (red) and Hoechst (blue). Arrowheads indicate DsRed+ CRs at P15 and P25. For simplicity, arrowheads were not displayed at P7, as there are too many CRs at this stage. (D) Quantification of CR density (CRs/mm3) at the pial surface in the somatosensory (S1) cortex (P7: n = 5 for controls and n = 6 for mutants, p=0.2251; P15: n = 4 for controls and n = 3 for mutants, p=0.771; P25: n = 3 for controls and n = 3 for mutants, p=0.813). Mann-Whitney U Test. Scale bar represents 200 μm. Data used for quantitative analyses as well as the numerical data that are represented in graphs are available in Figure 1—figure supplement 1—source data 1.

A similar proportion of CRs were rescued either by hyperpolarization (Figure 1A and B) or by preventing apoptosis (Ledonne et al., 2016), raising the intriguing possibility that the same subpopulation of CRs, SE-CRs, might be preserved in both conditions. Since hem-CRs are not eliminated via a Bax-dependent process (Ledonne et al., 2016), we investigated whether their survival is sensitive to hyperpolarization. We thus overexpressed Kir2.1 specifically in hem-CRs using the Wnt3acre/+, which corresponds to about 70% of the ΔNp73-CRs at early postnatal stages (Figure 1C and D). We found that hem-CR death was unaffected in this mouse line (Figure 1C and D). Moreover, taking into account cortical growth, the proportions of rescued cells in the somatosensory cortex at P25 was evaluated to approximately 30% of the initial pool of ΔNp73-CRs, which corresponds to the expected number of SE-CRs (Bielle et al., 2005; Yoshida et al., 2006; Tissir et al., 2009). Collectively, these results show that the death of a specific subset of ΔNp73cre/+ SE-CRs is both Bax-dependent and activity-dependent.

CRs rescued by hyperpolarization or blocking Bax-dependent apoptosis are integrated in neuronal circuits

It has been established that CRs are integrated in functional circuits early in the developing postnatal neocortex (Kilb and Luhmann, 2001; Soda et al., 2003; Sava et al., 2010; Cocas et al., 2016). CRs receive GABAergic synaptic inputs and, despite the expression of NMDA receptors (NMDARs) on CR membranes, the presence of NMDAR-mediated synaptic responses is still under debate (Kilb and Luhmann, 2001; Soda et al., 2003; Sava et al., 2010; Schwartz et al., 1998; Mienville and Pesold, 1999; Radnikow et al., 2002; Anstötz et al., 2014). To test whether CRs harbored functional GABAergic and/or glutamatergic synapses during the cell death period and after their rescue, we recorded spontaneous (sPSCs) and evoked postsynaptic currents (ePSCs) of fluorescent CRs with a KCl-based intracellular solution. First, at P9-11 in control ΔNp73cre/+;R26mt/+ mice, sPSCs sensitive to the GABAA receptor (GABAAR) antagonist SR95531 (10 µM) were observed in CRs held at −60 mV, confirming that CRs are innervated mainly by functional GABAergic synaptic inputs (Figure 2—figure supplement 1A–B; data not shown for SR95531 application; n = 7). Low-frequency stimulations in LI easily elicited ePSCs that were also completely blocked by SR95531 application (Figure 2—figure supplement 1C–D), even in 0 mM Mg2+ condition, which relieves the Mg2+ block of NMDA receptors. Taken together, these results indicate that ePSCs were mediated by GABAARs and not by AMPA or NMDA receptors. In agreement with previous studies (Sun et al., 2019), the presence of GABAergic synaptic inputs on CRs was confirmed by immunostainings against the presynaptic marker GAD65/67 and the postsynaptic marker Gephyrin (Figure 2—figure supplement 1E). Since the depolarizing action of GABA in CRs was proposed to be partly due to the lack of expression of KCC2, the chloride transporter responsible for maintaining a low intracellular chloride concentration (Mienville, 1998; Achilles et al., 2007; Pozas et al., 2008; Blanquie et al., 2017a; Blanquie et al., 2017b), we also checked for the protein expression of this transporter. In agreement with previous reports (Achilles et al., 2007; Pozas et al., 2008), control GFP+ CRs in ΔNp73cre/+;TauGFP/+ mice expressed very low to undetectable levels of KCC2 (Figure 2—figure supplement 1F). Thus, during the timeperiod of their activity-dependent death, CR cells receive solely GABAergic synaptic inputs.

We further explored whether these inputs are maintained at later stages, when CR cells are not eliminated. To this aim, we performed the same experiments at P23-28 in ΔNp73cre/+;R26Kir2.1/+ and ΔNp73cre/+;Baxlox/lox;R26mT/+ mice (Figure 2A–B). First, we found that sPSCs in rescued CRs had similar frequencies, amplitudes and kinetics in both models (Figure 2A–B). Moreover, sPSCs were completely abolished by GABAAR antagonist SR95531, indicating that rescued CRs remained innervated by functional GABAergic synaptic inputs (data not shown for SR95531 application, n = 8 and n = 5 for ΔNp73cre/+;R26Kir2.1/+ and ΔNp73cre/+;Baxlox/lox;R26mT/+ mice, respectively). Consistently, the complete block of ePSCs by SR95531, even in 0 mM Mg2+, revealed that ePSCs were exclusively mediated by GABAARs at hyperpolarized holding potentials, as observed in younger mice (Figure 2C–D and Figure 2—figure supplement 1C–D). While these GABAergic inputs were preserved in rescued CRs, we observed a reduced sPSC frequency and ePSC amplitude compared to P9-P11 mice, suggesting a decreased CR connectivity in the more mature neocortex (Figure 2A–B and Figure 2—figure supplement 1A–B).

Figure 2 with 1 supplement see all
Download asset Open asset
Pure GABAergic sPSCs and ePSCs in rescued CRs.

(A) Spontaneous PSCs (sPSCs) recorded in rescued CRs from ΔNp73cre/+;R26Kir2.1/+ at P27 (blue) and ΔNp73cre/+;Baxlox/lox mutants at P26 (green), respectively. (B) Plots of the frequency and amplitude of sPSCs (n = 11 for ΔNp73cre/+; R26Kir2.1/+ and n = 11 for ΔNp73cre/+; Baxlox/lox mice at P24-29; frequency: p=0.552, amplitude: p=0.580, Student T Test). Rise time is 2.10 ± 0.42 ms vs 1.02 ± 0.20 ms and decay time 34.26 ± 6.39 ms vs 29.14 ± 3.56 ms for ΔNp73cre/+;R26Kir2.1/+ and ΔNp73cre/+;Baxlox/lox mice, respectively. (C) Mean evoked PSCs (ePSCs) for rescued CRs respectively from a ΔNp73cre/+;R26Kir2.1/+ mutant at P29 (blue) and a ΔNp73cre/+;Baxlox/lox mutant at P26 (green) upon stimulation of LI neuronal fibers (stimulation time, arrowhead) in control conditions (top), with SR95531 (middle) and SR95531 in Mg2+-free solution (bottom). Note that ePSCs completely disappeared after bath application of SR95531. (D) Amplitudes of ePSCs in control conditions, with SR95531 and with SR95531 in Mg2+-free solution (ΔNp73cre/+;R26Kir2.1/+ mice at P24-29: ncontrol = 10, nSR95531 = 8 and nSR95531/Mg2+free=8; ΔNp73cre/+;Baxlox/lox: ncontrol = 8, nSR95531 = 5 and nSR95531/Mg2+free=5; Kruskal-Wallis test followed by a Bonferroni multiple comparison when comparing the three conditions for each mutant; Student T test for comparison of control ePSCs between ΔNp73cre/+;R26Kir2.1/+ and ΔNp73cre/+;Baxlox/lox mutants, p=0.638). To detect CRs in ΔNp73cre/+;Baxlox/lox mutants the R26mT/+ reporter line was used. Data used for quantitative analyses as well as the numerical data that are represented in graphs are available in Figure 2—figure supplement 1—source data 1.

Altogether, these results indicate that rescued CRs, as previously showed in the early postnatal neocortex (Soda et al., 2003; Sava et al., 2010; Kilb and Luhmann, 2001), receive solely GABAergic synaptic inputs at the time of their death. They further demonstrate that rescued CRs in both mouse models, inspite of a reduced connectivity compared to earlier stages, are kept integrated in functional neuronal circuits.

Survival of electrically-active CRs triggers dendritic exuberance of upper layer pyramidal neurons

In order to test whether CR aberrant survival may alter the function of other neurons in upper cortical layers, we used biocytin-filling and confocal 3D reconstructions to study the morphology of Layer II (LII) and LIII pyramidal neurons in the somatosensory barrel cortex of both ΔNp73cre/+;R26Kir2.1/+ and ΔNp73cre/+;Baxlox/lox mice. While no major changes were observed in the morphology of LII/III pyramidal neurons of ΔNp73cre/+;R26Kir2.1/+ mice compared to their matched controls (Figure 3—figure supplement 1), major defects were observed for these neurons in ΔNp73cre/+;Baxlox/lox mutants (Figure 3). Quantitative analyses revealed an increase in the number of apical and basal dendritic branches at P25 in ΔNp73cre/+;Baxlox/lox compared to controls (Figure 3B). In order to further characterize cell complexity in relation with the distance from the soma, we performed a Sholl analysis. We observed that LII/III pyramidal cells displayed an increased complexity for apical dendrites (between 180 µm and 240 µm from the soma) as well as for basal dendrites (between 60 and 80 µm from the soma) (Figure 3C). Interestingly, no statistically different changes in the number of dendritic branches (Figure 3—figure supplement 1B) or the complexity of apical and basal dendrites (Figure 3—figure supplement 1C) were detected in LII/LIII pyramidal neurons of the ΔNp73cre/+;R26Kir2.1/+ model, in which SE-CRs survive but are hyperpolarized. Thus, the survival of a specific subset of CRs has a general promoting impact onto the dendritic tree of upper layer pyramidal neurons only when they keep their normal intrinsic excitability. Collectively, our analyses show that SE-CRs survival in LI promotes an exuberance of apical and basal dendrites of LII/LIII pyramidal neurons in an activity-dependent manner.

Figure 3 with 1 supplement see all
Download asset Open asset
Increased dendritic branches in LII/LIII pyramidal neurons of ΔNp73cre/+;Baxlox/lox mutants.

(A) Representative examples of LII/III pyramidal neurons filled with biocytin in control (P25) and ΔNp73cre/+;Baxlox/lox mutant (P24) somatosensory cortex. (B) Quantification of the number of dentritic branches in control and ΔNp73cre/+;Baxlox/lox mutant LII/III pyramidal neurons, expressed as a percentage of dendritic branches relative to the mean of controls (n = 7 for controls and n = 8 for mutants at P23-28 p=0.0182 for apical dendrites and p=0.014 for basal dendrites; Mann-Whitney U Test). (C) Sholl analysis for the apical and basal dendrites in control and ΔNp73cre/+;Baxlox/lox mutants showing an increased cell complexity between 180 and 240 µm (p-value=0.04, 0.027, 0.0007 and 0.005, respectively) and 60 and 80 µm (p value=0.027 and 0.019, respectively) from the soma, respectively (n = 7 for controls and n = 8 for mutants). Multiple T-test. Scale bar represents 100 μm. Data used for quantitative analyses as well as the numerical data that are represented in graphs are available in Figure 3—figure supplement 1—source data 1.

Survival of electrically-active CRs increases excitatory entries in upper layer pyramidal neurons

To examine whether the defects of the dendritic arborization of LII/III pyramidal neurons were related to changes in the synaptic inputs received by these neurons, we first examined spines on both apical and basal dendrites of biocytin-filled pyramidal cells (Figure 4A–D). Because excitatory synapses are formed on dendritic spines, the latter can be used as a proxy for the quantification of those synapses. For apical dendrites, we examined terminal ramifications in LI, whereas for basal dendrites, we considered horizontal branches approximately at the same distance from the soma. Spine density on both apical and basal dendrites of pyramidal cells was significantly increased in ΔNp73cre/+;Baxlox/lox mutants compared to controls (Figure 4A and B) whereas no differences were observed in ΔNp73cre/+;R26Kir2.1/+ mutants (Figure 4C and D). These results indicate that the survival of electrically-active CRs not only triggers a dendritic exuberance, but also drives an increase in spine densities.

Figure 4 with 1 supplement see all
Download asset Open asset
Spine density and evoked synaptic activity recorded in LII/LIII pyramidal neurons in both ΔNp73cre/+;Baxlox/lox and ΔNp73cre/+;R26Kir2.1/+ mutants.

(A, C) Representative confocal images showing spines in apical and basal dendritic segments of controls (left) and ΔNp73cre/+;Baxlox/lox (A, right) and ΔNp73cre/+;R26Kir2.1/+ mutants (C, right) at P24-25. (B, D) Quantification of the spine density (number of spines/µm) in apical and basal dendrites in LII/LIII pyramidal neurons for both controls and mutants from the same litters (for ΔNp73cre/+;Baxlox/lox apical and basal dendrites: n = 8 for controls and n = 12 for mutants at P23-28, p=0.012 for apical dendrites and p=0.0014 for basal dendrites; for ΔNp73cre/+;R26Kir2.1/+ apical dendrites: n = 10 for controls and n = 6 for mutants at P23-P29, p=0.166; basal dendrites: n = 9 for controls and n = 7 for mutants, p=0.652; Mann-Whitney U Test). Scale bar represents 5 μm. (E, G) Pyramidal neurons recorded in voltage-clamp at −70 mV and 0 mV in control at P26 (E, left) and P24 (G, left) and in a ΔNp73cre/+;Baxlox/lox mutant at P23 (E, right) and a ΔNp73cre/+;R26Kir2.1/+ mutant at P28 (G, right) during the extracellular stimulation of LII/III fibers as indicated (E, inset). Stimulation artefacts were blanked for visibility. The stimulation time is indicated (arrowheads). (F, H) Plots of E/I ratio calculated from eEPSCs and eIPSCs in controls and ΔNp73cre/+;Baxlox/lox mutants (F) and ΔNp73cre/+; R26Kir2.1/+ mutants (H) (for ΔNp73cre/+;Baxlox/lox: n = 10 for controls and n = 14 for mutants, p=0.031, Student T Test; for ΔNp73cre/+;R26Kir2.1/+: n = 8 for controls and n = 7 for mutants, p=0.612; Mann-Whitney U Test). Data used for quantitative analyses as well as the numerical data that are represented in graphs are available in Figure 4—figure supplement 1—source data 1.

To test whether these morphological modifications in ΔNp73cre/+;Baxlox/lox mutants is accompanied by modifications in excitatory synaptic inputs, we performed whole-cell recordings of upper layer pyramidal neurons during the extracellular stimulation of LII/III fibers. First, the membrane potential of recorded cells was maintained at −70 mV or 0 mV to respectively record evoked excitatory (eEPSCs) and inhibitory (eIPSCs) postsynaptic currents. Pyramidal cells in ΔNp73cre/+;Baxlox/lox mutants showed a significant increase in the mean amplitude of eEPSCs, while that of eIPSCs remained unchanged compared to controls (Figure 4E, Figure 4—figure supplement 1A). This modification is highlighted by a significant increase in the E/I ratio (Figure 4F). Together with the increased spine density, these data strongly suggest that pyramidal neurons have enhanced excitatory synaptic inputs. To corroborate this possibility, we then analyzed the spontaneous EPSCs (sEPSCs) of recorded pyramidal neurons. As expected for an increased number of inputs, the sEPSC frequency was significantly higher in ΔNp73cre/+;Baxlox/lox mutants with respect to controls while the mean sEPSC amplitude remained unchanged (Figure 4—figure supplement 1C-D). When the same experiments were performed in ΔNp73cre/+;R26Kir2.1/+ mutants, no differences were observed either in the amplitudes of both eEPSCs and eIPSCs or in the E/I ratio (Figure 4G–H, Figure 4—figure supplement 1B). In line with this, changes were neither observed in the frequency of sEPSCs of these mutants (Figure 4—figure supplement 1E), showing that defects in the synaptic activity of pyramidal neurons is dependent on the intrinsic activity of rescued CRs. To test whether the effect on pyramidal cells could be due to a direct action of CR activity on excitatory circuits, we produced ΔNp73cre/+;Baxlox/lox; ChR2lox/+ mutant mice to photoactivate rescued CRs while recording neuronal network activity. After defining an efficient photostimulation protocol for reliably eliciting action potentials on recorded CRs (Figure 4—figure supplement 1F), we combined photostimulation with of Layer I interneuron recordings in whole-cell configuration and/or local field potentials (LFPs) in different layers (Figure 4—figure supplement 1G). We could never detect light-evoked responses during patch-clamp or extracellular recordings, even in the presence of 0 mM Mg2+, 3 mM Ca2+ and 50 µM of the potassium channel blocker 4AP, a treatment that renders all neurons more excitable (Figure 4—figure supplement 1G). Although these results do not rule out that rescued CRs contact other neurons through bona fide glutamatergic synapses, the low proportion of these cells compared to pyramidal neurons and the lack of electrical extracellular responses during their sustained light stimulation suggest that direct synaptic inputs from CRs cannot account for the robust morphological and functional changes induced in pyramidal cells by the aberrant CR survival.

Overall, these experiments show that the survival of electrically-active SE-CRs increases the excitatory inputs to upper pyramidal neurons, possibly through a non-glutamatergic mechanism, thereby generating a E/I imbalance and functional changes in circuit wiring.

Discussion

Our results show that the elimination of specific subsets of CRs, SE-CRs, is activity-dependent and that this process is essential for proper cortical wiring. Indeed, the persistence of SE-CRs beyond their normal phase of elimination triggered major deficits in LII/LIII somatosensory pyramidal neurons. Not only the analyzed neurons displayed an increased dendritic arborization and spine density, but they also consistently showed enhanced excitatory inputs leading to a functional E/I imbalance. Remarkably, these anatomical and electrophysiological deficits all relied on the fact that persistent CRs were electrically-active. Our study thus demonstrates that activity is required to eliminate SE-CRs, whose survival would otherwise perturb cortical wiring in an activity-dependent manner. Taken together, it reveals an elegant interplay between transient CRs and neuronal activity in the construction of functional cortical excitatory circuits.

CR subtype-specific pathways in programmed cell death

Activity was reported to promote survival of both glutamatergic and GABAergic neurons in the neocortex and in general in the nervous system ( Blanquie et al., 2017a; Causeret et al., 2018; Wong and Marín, 2019), providing a mean to integrate around 70% of neurons into functional circuits. CRs, which completely undergo programmed cell death in the cerebral cortex (Ledonne et al., 2016; Causeret et al., 2018), have been previously proposed to behave differently. These neurons display ‘immature’ features such as a depolarized resting potential and a very high input resistance (Kirischuk et al., 2014). Especially, GABA is depolarizing in these cells due to their elevated intracellular chloride concentration resulting from the activity of the chloride inward transporter NKCC1 in the absence of expression of chloride outward transporter KCC2 (Mienville, 1998; Achilles et al., 2007; Pozas et al., 2008). Interestingly, the pharmacological blockade of NKCC1 in cell cultures or the genetic ablation of this transporter in vivo promotes the survival of a CR population, probably by preventing GABAA receptor-mediated depolarization (Blanquie et al., 2017b). It must be considered, however, that the depolarizing effect of GABA will depend most probably on the levels of neuronal activity since high activity levels attenuate the GABAA receptor-mediated excitatory drive in CRs (Kolbaev et al., 2011). Here, we confirmed that CRs receive exclusively functional GABAergic synaptic inputs and express low levels of KCC2 in the second postnatal week, that is during the period of massive cell death. Our findings raise the question of the identity of possible GABAergic neurons that regulate CR subtype elimination. CRs receive GABAergic synaptic inputs from different sources, including local interneurons of Layer I, the underlying layers of the neocortex and the subplate, a transient cortical structure absent in the fourth postnatal week, as well as the zona incerta (Kirmse et al., 2007; Myakhar et al., 2011; Kirischuk et al., 2014; Chen and Kriegstein, 2015; Sun et al., 2019). A related issue is how GABAergic inputs to CRs might change overtime, since we found that rescued CRs receive less inputs than at earlier stages. One interesting possibility is that rescued CRs in Layer I might lose GABAergic innervation from transient or distant populations (i.e. subplate and zona incerta) during maturation of neuronal networks, restricting their connectivity with more local neocortical inputs. Further investigation is needed to determine the different interneuron subtypes impinging on CRs in immature cortical circuits and after their aberrant survival in adults.

Using conditional Kir2.1 expression in a large subpopulation of CRs, we unequivocally show that only a specific subset of CRs, SE-CRs, dies in an activity-dependent manner in vivo. Since CRs are highly hyperpolarized in this mouse model, a plausible explanation is that GABAergic inputs cannot exert their depolarizing effect as in normal conditions, thereby preventing cell death. In this context, it is also tempting to hypothesize that, as shown in other systems, neuronal activity via intracellular calcium signals could trigger apoptosis, as reported during excitotoxicity (Blanquie et al., 2017a). The timing of SE-CRs death, namely the second postnatal week, corresponds to a major switch in cortical activity (Luhmann and Khazipov, 2018) and in GABAergic circuits (Cossart, 2011), raising the possibility that the elimination of CRs is part of a more global activity-dependent remodeling of cortical circuits. Irrespective of the underlying mechanism, surviving Kir2.1-expressing CRs displayed a relatively normal morphology, expressed Reln and received GABAergic synaptic inputs. Remarkably, SE-CRs is also the subpopulation undergoing a Bax-dependent apoptosis. Quantification of CRs that persist in both models suggests that a vast majority of SE-CRs survives. Indeed, in both deletion of Bax (Ledonne et al., 2016) or over-expression of Kir2.1 (this manuscript) models, a five-fold increase in CR numbers, corresponding to approximately 30% of the initial pool at P7, is detected when using the ΔNp73Cre line, in contrast to none when using the hem-specific Wnt3aCre line. Since hem-CRs constitute about 70% of the population targeted by the ΔNp73Cre line (Bielle et al., 2005; Yoshida et al., 2006; Tissir et al., 2009), our findings support that the 30% of rescued CRs in ΔNp73Cre corresponds to a large fraction, if not all, of the SE-CR population. Together, these results demonstrate that hem-derived CRs die in a Bax- and activity-independent manner, in contrast to SE-CRs that survive in Bax and Kir2.1 mutants. Hence, our work reveals that subpopulations of CRs are eliminated by very distinct mechanisms. Interestingly, hippocampal CRs, which mostly derive from the cortical hem (Louvi et al., 2007) display a delayed death which seems to occur independently of the apoptotic-specific Caspase-3 activity (Anstötz et al., 2016; Anstötz et al., 2018). Together, these data support the notion that CR subtypes are intrinsically different in the mechanism determining their demise and argues in favor of complex yet unappreciated subtype-specific pathways leading ultimately to cell death.

CRs aberrant survival perturbs the morphology and connectivity of upper layer neurons in an activity-dependent manner

In this work, we have demonstrated that SE-CRs persistence in mice has a strong effect on LII/III pyramidal neuron morphology and excitatory circuits. Notably, in the ΔNp73cre/+;Baxlox/lox mutants, we observed an impact on both apical and basal dendrites. While the effect on apical dendrites could be direct via local surviving SE-CRs, the impact on basal ones might be circuit-mediated since excitatory entries in apical dendritic tufts were shown to modify basal dendrite synaptic plasticity (Williams and Holtmaat, 2019). Moreover, SE-CRs survival leads in ΔNp73cre/+;Baxlox/lox mutants to increased synaptic density with major functional consequences on the E/I ratio. Conversely, recent work showed that aberrant reduction of CRs during development triggers decreased apical dendritic tufts and dendritic spine density of LII/III pyramidal neurons accompanied by a reduction in the E/I ratio (de Frutos et al., 2016). Taken together our findings reveal that the proper balance of CRs constitutes an essential, yet underappreciated, regulator of LII/III pyramidal neuron morphology and wiring.

Importantly, we found in both Bax and Kir models that surviving CRs are similarly kept embedded into functional networks. Interestingly, rescued CRs are also solely innervated by GABAA receptor-mediated synapses like their younger control counterparts. Although GABAergic synaptic connectivity often increases during postnatal development (Pangratz-Fuehrer and Hestrin, 2011), some cells may display transient connections that disappear after the second postnatal week, thereby accounting for the reduced connectivity observed in the two models. In contrast, since the effects on pyramidal neurons and cortical excitability are only found in Bax mutants, our study reveals that the inappropriate survival of SE-CRs drives an abnormal cortical wiring via an activity-dependent mechanism. However, CRs appear to act on upper layer pyramidal neurons via partially distinct mechanisms at different time points during development. Indeed, Kir2.1 expressing mice appear largely similar to controls, suggesting that the impact of reducing CR density on LII/III apical dendrites (de Frutos et al., 2016) does not rely exclusively on the intrinsic excitability of CRs. Nevertheless, the mechanism by which rescued CRs induce morphological and functional changes on pyramidal neurons remains unresolved. The lack of response observed during our optogenetic experiments suggest that these changes may not depend on a direct CR excitatory synaptic input onto principal neurons. This is in line with recent data obtained with experiments performed in the hippocampus using ChR2 activation and paired-recordings in the third postnatal week, where CRs are still present in high density (Quattrocolo and Maccaferri, 2014; Anstötz et al., 2016). Indeed, only very few pyramidal cells could be detected as an output of CR cells. Further experiments will be required to determine the relative roles of CR-secreted factors versus circuit-mediated effects onto apical and basal dendrites. Regardless the mechanism and since excitatory entries onto apical dendrites are emerging as major actors in sensory gating, cortical integration and reward (Keller and Mrsic-Flogel, 2018; Khan and Hofer, 2018; Lacefield et al., 2019; Williams and Holtmaat, 2019; Zhang and Bruno, 2019), our findings highlight the importance of a transient cell population in the emergence of functional circuits as well as the deleterious effects of their abnormal demise.

Our study thus shows that the elimination of SE-CRs in the somatosensory cortex is required for proper morphology and wiring of LII/III pyramidal neurons and reveals a remarkable interplay between activity, the elimination of transient cells and cortical wiring. Indeed, activity, likely driven by cortical maturation, regulates the elimination of transient CRs that would otherwise perturb upper layer wiring. Notably, CRs persistence has been described in human pathological conditions, often associated with epilepsy. Our work thus not only provides novel insights onto normal wiring of upper layers, but also addresses the functional consequences of incomplete CR removal with major relevance for neurodevelopmental diseases, such as autism spectrum disorder, schizophrenia or epilepsy.

Materials and methods

Key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Strain Mus musculus (males and female)C57BL6JJanvier
Mus musculus (males and female)ΔNp73CreIRESGFPTissir et al., 2009ΔNp73Cre
Mus musculus (males and female)Wnt3aCreYoshida et al., 2006Wnt3aCre
Mus musculus (males and female)TauloxP-stop-loxP-MARCKSeGFP-IRES-nlslacZHippenmeyer et al., 2005TauGFP
Mus musculus (males and female)Baxtm2Sjk;Bak1tm1Thsn/JTakeuchi et al., 2005Baxlox/lox
Mus musculus (males and female)ROSA26loxP-stop-loxP-TomatoMadisen et al., 2010R26mT
Mus musculus (males and female)ROSA26 loxP-stop-loxP- Kcnj2-cherry/+Moreno-Juan et al., 2017R26Kir2.1/+
Mus musculus (males and female)Ai32(RCL-ChR2(H134R)/EYFPhttps://www.jax.org/strain/012569ChR2lox
Antibodyrabbit polyclonal anti-DsRedTakaraRRID:AB_10013483IF(1:500)
Antibodymouse monoclonal anti-ReelinMerck MilliporeRRID:AB_565117IF(1:300)
Antibodymouse monoclonal anti-GephyrinSynaptic systemsRRID:AB_2619837IF(1:250)
Antibodyrabbit polyclonal anti-GAD65/67MerckRRID: AB_22
78725		IF(1:250)
Antibodyguinea-pig anti-KCC2D Ng and S Morton
TM Jessell’s lab		IF(1:4000)
Antibodydonkey anti-mouse
Alexa-488		Jackson ImmunoResearch LaboratoriesRRID:AB_2340846IF(1:800)
Antibodydonkey anti-rabbit
Cy3		Jackson ImmunoResearch LaboratoriesRRID:AB_2307443IF(1:800)
Antibodydonkey anti-rabbit Alexa-647Molecular ProbesRRID:AB_2536183IF(1:500)
Antibodydonkey anti-chick Alexa-488Jackson ImmunoResearch LaboratoriesRRID:AB_2340375IF(1:1000)
Antibodydonkey anti-mouse Alexa-555Molecular ProbesRRID:AB_2536180IF(1:1000)
Antibodygoat anti-guinea pig Alexa-555Molecular ProbesRRID:AB_2535856IF(1:1000)
AntibodyDAPI (4', 6-diamidino-2-phenylindole)Invitrogen Molecular ProbesRRID:AB_2629482IF(1:2000)
AntibodyDyLight 488 streptavidinVector LabsSP-4488
Sequence-based reagentCRE genotyping
188 f
167 r		This paperPCR primers188 f: TGA TGG ACA TGT TCA GGG ATC
167 r: GAA ATC AGT GCG TTC GAA CGC TAG A
Sequence-based reagentR26Kir2.1/+genotyping
AAY101
AAY103
SD297		PCR primersAAY101: AAAGTCGCTCTGAGTTGTTAT (Rosa26 forward WT)
AAY103: GGGAGCGGGAGAAATGGATATG (Rosa26 reverse WT)
SD297: GGCCATTTACCGTAAGTTATG (CAG promoter reverse)
Chemical compound, drugParaformaldehydeSigma-AldrichCAT:P6148
Chemical compound, drugTriton 100XEurobioCAT:GAUTTR00-07
Chemical compound, drugSR95531AbcamAb120042
Chemical compound, drug4-APSigma AldrichA-0152
Software, algorithmIMARIS software 8.4.IMARISRRID:SCR_007370
Software, algorithmGraphPad Prism 7.0GraphPad
Software		RRID:SCR_000306
Software, algorithmImageJ/FIJINIHRRID:SCR_002285
Software, algorithmAdobe
Photoshop CS6		Adobe
Systems		RRID:SCR_014199
Software, algorithmpClamp10.1Molecular DevicesRRID:SCR_011323
Software, algorithmIGOR Pro 6.0WavemetricsRRID:SCR_000325
Software, algorithmNeuroMaticWavemetricsRRID:SCR_004186

Animals

ΔNp73CreIRESGFP(ΔNp73Cre) (Tissir et al., 2009), Wnt3aCre (Yoshida et al., 2006), ROSA26loxP-stop-loxP-Tomato(R26mT) (Madisen et al., 2010), TauloxP-stop-loxP-MARCKSeGFP-IRES-nlslacZ (TauGFP) (Hippenmeyer et al., 2005) and ChR2lox (Ai32(RCL-ChR2(H134R)/EYFP) (https://www.jax.org/strain/012569​)) transgenic mice were kept in a C57BL/6J background. The Baxtm2Sjk;Bak1tm1Thsn/J line (Takeuchi et al., 2005) harboring the floxed Bax and the Bak knock-out alleles was purchased from the Jackson laboratory as mixed B6;129. ΔNp73Cre and Wnt3aCre lines were crossed with the R26mT or TauGFP reporter lines to permanently label CR subtypes. ΔNp73Cre line was crossed to the Baxtm2Sjk;Bak1tm1Thsn/J line (Baxlox/lox) to inactivate Bax function in specific CR subtypes. The conditional knock-out also harbored a reporter allele R26mT in order to trace the neurons in which recombination had occurred. ΔNp73Cre and Wnt3aCre lines were crossed to ROSA26 loxP-stop-loxP- Kcnj2-cherry/+ (R26Kir2.1/+) (Moreno-Juan et al., 2017) to overexpress the Kcnj2 gene, which encodes Kir2.1, in specific CR subpopulations. Controls used were littermates heterozygous Bax for the Bax model and ΔNp73+/+;R26Kir2.1/+ for the Kir2.1 model. Animals were genotyped by PCR using primers specific for the different alleles. All animals were handled in strict accordance with good animal practice as defined by the national animal welfare bodies, and all mouse work was approved by the Veterinary Services of Paris (Authorization number: 75–1454) and by the Animal Experimentation Ethical Committee Buffon (CEEA-40) (Reference: CEB-34–2012) and by the Animal Experimentation Ethical Committee Darwin (Reference: 02224.02).

Tissue preparation and immunohistochemistry

Request a detailed protocol

For staging of animals, the birth date was considered as postnatal day 0 (P0). Animals were anesthetized with Isoflurane and intracardially perfused with 4% paraformaldehyde (PFA) in 0.1 M PBS, pH 7.4 and post-fixed over-night in 4% PFA at 4°C. Brain were embedded in 3.5% agarose and sectioned in 70 µm free-floating slices at all stages in Figure 1 and Figure 1—figure supplement 1. For Figure 2—figure supplement 1E, brains were cryoprotected and sectioned in 50 µm free-floating slices. Immunostaining was performed as previously described (Bielle et al., 2005; Griveau et al., 2010; de Frutos et al., 2016). Primary antibodies used for immunohistochemistry were: mouse anti-Reelin (MAB5364, Millipore 1:300), rabbit anti-DsRed (Takara 632496, 1:500), mouse anti-Gephyrin (147 011, Synaptic Systems, 1:250), Rabbit anti-GAD65/67 (AB1511, Merck, 1:250), guinea pig anti-KCC2 (Gift of S.Morton and D.Ng, 1:4000). Secondary antibodies used against primary antibodies were: donkey anti-mouse Alexa-488 (Jackson ImmunoResearch Laboratories, 1:800), donkey anti-rabbit Cy3 (Jackson ImmunoResearch Laboratories, 1:800), donkey anti-chick Alexa-488 (Jackson ImmunoResearch Laboratories, 1:1000), donkey anti-mouse Alexa-555 (A-31570, Molecular Probes, 1:1000), donkey anti-rabbit Alexa-647 (A31573, Molecular Probes, 1:500), goat anti-guinea pig Alexa-555 (A-21435, Molecular Probes, 1:1000). Hoechst (Sigma-Aldrich 33342, 1:1000) and DAPI (D1306, ThermoFisher Scientific, 1:2000) were used for fluorescent nuclear counterstaining the tissue and mounting was done in Vectashield (Vector Labs).

Image acquisition and cell countings

Request a detailed protocol

Immunofluorescence images were acquired using a confocal microscope (Leica TCS SP5), except for anti-KCC2, Gephyrin and GAD65/67 (Figure 2—figure supplement 1E, f) that were acquired on a LEICA SP8 confocal microscope with 93X objective and 2.5 digital zoom (a single optical plane or around 100 z-stacks of 0.07 µm respectively). DsRed+ neurons, detected by immunofluorescence, were counted using the ImageJ software, in the somatosensory barrel cortex (S1) for each age and genotype. For each section, the density of CRs (DsRed+ CRs/mm3) was calculated taking into account the thickness of the section and the surface of Layer I, measured using ImageJ software.

Acute slice preparation, electrophysiology and photostimulation

Request a detailed protocol

Acute coronal slices (300 µm) of the neocortex were obtained from ΔNp73cre/+;Baxlox/lox and ΔNp73cre/+;R26Kir2.1/+ mutants. Excitation light to visualize the tdTomato or Cherry fluorescent proteins was provided by a green Optoled Light Source (Cairn Research, UK) and images were collected with an iXon+ 14-bit digital camera (Andor Technology, UK), as previously described (Orduz et al., 2015). Patch-clamp recordings were performed at RT using an extracellular solution containing (in mM): 126 NaCl, 2.5 KCl, 1.25 NaH2PO4, 26 NaHCO3, 20 glucose, five pyruvate, 2 CaCl2 and 1 MgCl2 (95% O2, 5% CO2). Fluorescent CRs were recorded at P13-17 and P24-29 with different intracellular solutions according to the experiment and containing (in mM): either 130 K-Gluconate (K-Glu) or 130 KCl, 0.1 EGTA, 0.5 CaCl2, 2 MgCl2, 10 HEPES, 2 Na2-ATP, 0.2 Na-GTP and 10 Na2-phosphocreatine and 5.4 mM biocytin (pH ≈ 7.3). When using a KGlu-based intracellular solution in whole-cell configuration, potentials were corrected for a junction potential of −10 mV. Recordings were made without series resistance (Rs) compensation; Rs was monitored during recordings and cells showing a change of more than 20% in Rs were discarded. To evaluate the effect of SR95351 (10 µM; Abcam, Cambridge, UK), drug perfusion reached a steady state at 3 min in the recording chamber. This time was respected before quantification of either spontaneous or evoked PSCs. To test for the presence of eEPSCs mediated by NMDARs receptors at hyperpolarized potentials, the extracellular concentration of MgCl2 was replaced by CaCl2 in order to relieve the Mg2+ block of these receptors.

Photostimulation of fluorescent ChR2-expressing rescued CRs recorded in whole-cell configuration was obtained by triggering light trains with a blue LED (470 nm, 1 ms pulses; Optoled Light Source, Cairn Research, UK). Light trains of 2, 5, 10 and 20 Hz during 10 s or 30 s were applied to define the optimal frequency inducing an effective activation of recorded ChR2-expressing rescued CRs. For each frequency, we calculated the number of spikes (Ns) with respect to the number of light pulses (NLP) and determined the percentage of success as [Ns/NLP] x 100. To test the effect of rescued CRs activation, we performed extracellular recordings and patch-clamp recordings of Layer I interneurons while stimulating with light trains (5 Hz, 10 s or 30 s). Extracellular recordings were recorded with a patch pipette filled with extracellular solution. In a set of experiments, the extracellular solution contained 0 mM Mg2+, 3 mM Ca2+ and 4AP for more than 5 min at least.

Recordings were obtained using Multiclamp 700B and pClamp10.1 (Molecular Devices), filtered at 4 kHz and digitized at 20 kHz. Digitized data were analyzed off-line using Neuromatic within IGOR Pro 6.0 environment (Wavemetrics, USA) (Rothman and Silver, 2018). Extracellular stimulations were performed using a monopolar electrode (glass pipette) placed in Layer I for CRs and Layer II/III for pyramidal neurons (20–99 V, 100 µs stimulations each 8–12 s; Iso-Stim 01D, npi electronic GmbH, Tamm, Germany). Spontaneous postysynaptic currents were detected with a threshold of 2 times the noise standard deviation during a time window of 3 min for CRs and 1.5 min for pyramidal cells. The Vm was estimated in current-clamp mode as soon as the whole-cell configuration was established. The analysis of Rin, action potential amplitudes and duration was performed during pulses of 800 ms in current-clamp configuration from −80 mV during increasing steps of 5 pA as previously described (Ledonne et al., 2016).

Morphological analyses

Request a detailed protocol

For morphological analysis, CRs and layer II/III pyramidal cells were loaded with biocytin through patch pipette during whole-cell recordings. The slices were fixed 2 hr in 4% paraformaldehyde at 4°C, rinsed three times in PBS for 10 min, and incubated with 1% triton X-100% and 2% BSA during 1 hr. Then, they were washed three times in PBS and incubated in DyLight 488 streptavidin (Vector Labs, Burlingame, USA) for 2 hr. Successfully labeled CRs and Layer II/III pyramidal cells were visualized either using a LEICA SP5 or SP8 confocal microscope with a 40X objective and a 1.3 digital zoom. Around 250 optical sections of 0.6 μm were necessary to image the whole dendritic tree of each neuron. For dendritic spines, images of apical dendrites in Layer I and basal dendrites in Layer II/III were acquired with a 63X objective (100 optical sections of 0.25 μm each). For apical dendrites, a terminal ramification in Layer I was acquired, while for basal dendrites a horizontal dendrite in Layer II/III was acquired approximately at 60 μm distance from the soma after the first ramification. 3D reconstruction was performed using the IMARIS software 8.4. Statistical analyses were performed based on the data given by IMARIS in apical and basal dendrites separately. Counting of spines was performed manually using ImageJ software on black and white maximum projections on two different segments approximately 50 µm long for each image. For morphological analysis of CRs, statistical analyses were performed based on the values calculated by image analyses using the IMARIS software (soma diameter and filament length).

Statistical analysis

Request a detailed protocol

All data were expressed as mean ± SEM. A P-value less than 0.05 was considered significant. For statistical groups larger than 7, we performed a D’Agostino-Pearson normality test. According to the data structure, two-group comparisons were performed using two-tailed unpaired Student T test or Mann-Whitney U Test. Bonferroni multiple comparisons were used as post-hoc test following one-way or two-way ANOVA or the non-parametric Kruskal-Wallis Tests. For small statistical groups (less than 7), we systematically performed non-parametric tests (Mann-Whitney U Test or Kruskal-Wallis Tests with Dunn’s correction). A reconstructed pyramidal neuron displaying an exceptional basal dendrite projecting to Layer I was submitted to Grubbs test and excluded as an extreme outlier. Statistics and plotting were performed using GraphPad Prism 7.00 (GraphPad Software Inc, USA). *p<0.05, **p<0.01, ***p<0.001.

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files. The Source data file contains all the data presented in the figures (1 sheet per Figure).

References

  29. Book
    1. Luhmann HJ
    (2013) Cajal-Retzius and Subplate cells: Transient Cortical neurons and Circuits
    In: Rubenstein J. L. R, Rakic P, editors. Cellular Migration and Formation of Neuronal Connections: Comprehensive Developmental Neuroscience.  Academic Press. pp. 843–856.
    https://doi.org/10.1016/B978-0-12-397266-8.00009-0

Article and author information

Author details

  1. Martina Riva

    1. Institut Imagine, Université de Paris, Paris, France
    2. Institut Jacques Monod, CNRS UMR 7592, Université de Paris, Paris, France
    3. Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Université de Paris, Paris, France
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    Contributed equally with
    Ioana Genescu and Chloé Habermacher
    Competing interests
    No competing interests declared

  2. Ioana Genescu

    Institut de Biologie de l’École Normale Supérieure (IBENS), Département de Biologie, École Normale Supérieure, CNRS, INSERM, Université PSL, Paris, France
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    Contributed equally with
    Martina Riva and Chloé Habermacher
    Competing interests
    No competing interests declared

  3. Chloé Habermacher

    1. Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Université de Paris, Paris, France
    2. INSERM U1128, Paris, France
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    Contributed equally with
    Martina Riva and Ioana Genescu
    Competing interests
    No competing interests declared

  4. David Orduz

    INSERM U1128, Paris, France
    Present address
    Gfi informatique, Saint-Ouen-sur-Seine, France
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-4198-2691

  5. Fanny Ledonne

    Institut Jacques Monod, CNRS UMR 7592, Université de Paris, Paris, France
    Contribution
    Data curation, Formal analysis, Supervision, Validation, Visualization, Methodology
    Competing interests
    No competing interests declared

  6. Filippo M Rijli

    Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland
    Contribution
    Resources
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0515-0182

  7. Guillermina López-Bendito

    Instituto de Neurosciencias de Alicante, Universidad Miguel Hernandez, Sant Joan d’Alacant, Spain
    Contribution
    Resources
    Competing interests
    No competing interests declared

  8. Eva Coppola

    1. Institut Imagine, Université de Paris, Paris, France
    2. Institut Jacques Monod, CNRS UMR 7592, Université de Paris, Paris, France
    3. Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Université de Paris, Paris, France
    Contribution
    Conceptualization, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing—original draft, Writing—review and editing
    Competing interests
    No competing interests declared

  9. Sonia Garel

    Institut de Biologie de l’École Normale Supérieure (IBENS), Département de Biologie, École Normale Supérieure, CNRS, INSERM, Université PSL, Paris, France
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing
    Contributed equally with
    Maria Cecilia Angulo and Alessandra Pierani
    For correspondence
    garel@biologie.ens.fr
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-2984-3645

  10. Maria Cecilia Angulo

    1. Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Université de Paris, Paris, France
    2. INSERM U1128, Paris, France
    Contribution
    Conceptualization, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing
    Contributed equally with
    Sonia Garel and Alessandra Pierani
    For correspondence
    maria-cecilia.angulo@parisdescartes.fr
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-0758-0496

  11. Alessandra Pierani

    1. Institut Imagine, Université de Paris, Paris, France
    2. Institut Jacques Monod, CNRS UMR 7592, Université de Paris, Paris, France
    3. Institute of Psychiatry and Neuroscience of Paris (IPNP), INSERM U1266, Université de Paris, Paris, France
    Contribution
    Conceptualization, Resources, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing
    Contributed equally with
    Sonia Garel and Maria Cecilia Angulo
    For correspondence
    alessandra.pierani@inserm.fr
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-4872-4791

Funding

Agence Nationale de la Recherche (ANR-15-CE16-0003-01)

  • Alessandra Pierani
  • Maria Cecilia Angulo
  • Sonia Garel

Fondation pour la Recherche Médicale (Equipe (DEQ20130326521))

  • Alessandra Pierani

Fondation pour la Recherche Médicale (Equipe (DEQ20150331681))

  • Maria Cecilia Angulo

European Commission (ERC-2013-CoG-616080)

  • Sonia Garel

Ministry of Higher Education, Research and Innovation (Fellowship)

  • Martina Riva
  • Ioana Genescu

Ministry of Science, Innovation and Universities (BFU2015-64432-R)

  • Guillermina López-Bendito

European Commission (ERC-2014-CoG-647012)

  • Guillermina López-Bendito

Fondation pour l'Aide à la Recherche sur la Sclérose en Plaques (Postdoctoral fellowship)

  • Chloé Habermacher

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

The authors apologize for not having been able to cite the work of many contributors to the field. We wish to thank Q Dholandre, L Vigier, M Keita, D Souchet, C Auger, A Delecourt, E Touzalin, D Valera and C Le Moal, for help with the mouse colonies and genotyping, L Danglot, S Morton and D Ng for providing antibodies and A Ben Abdelkrim for help with statistical analyses, as well as members of the Pierani, Angulo and Garel laboratories for helpful discussions and critical reading of the manuscript. We acknowledge the ImagoSeine facility, member of the France BioImaging infrastructure supported by the French National Research Agency (ANR-10-INSB-04, ‘Investments for the future’) for help with confocal microscopy, Animalliance for technical assistance and animal care. We thank the IBENS Imaging Facility (France BioImaging, supported by ANR-10-INBS-04, ANR-10-LABX-54 MEMO LIFE and ANR-11-IDEX-000–02 PSL* Research University, ‘Investments for the future’). We acknowledge NeurImag facility of IPNP. AP and MCA are CNRS (Centre National de la Recherche Scientifique) Investigators, SG is an Inserm researcher, and all member Teams of the École des Neurosciences de Paris Ile-de-France (ENP), EC is a University Paris Diderot Lecturer, MR and IG are supported by fellowships from the French Ministry of Research, CH by a postdoctoral fellowship from Fondation pour l'aide à la recherche sur la Sclérose en Plaques (ARSEP). This work was supported by grants from the ANR-15-CE16-0003-01, FRM («Equipe FRM DEQ20130326521») to AP and State funding from the Agence Nationale de la Recherche under ‘Investissements d’avenir’ program (ANR-10-IAHU-01) to the Imagine Institute, Fondation pour la Recherche Médicale (FRM, «Equipe FRM DEQ20150331681») to MCA, grants from INSERM, CNRS and the ERC Consolidator Grant NImO 616080 to SG, ERC Consolidator Grant (ERC-2014-CoG-647012) and the Spanish Ministry of Science, Innovation and Universities (BFU2015-64432-R) to GL-B.

Ethics

Animal experimentation: All animals were handled in strict accordance with good animal practice as defined by the national animal welfare bodies, and all mouse work was approved by the Veterinary Services of Paris (Authorization number: 75-1454) and by the Animal Experimentation Ethical Committee Buffon (CEEA-40) (Reference: CEB-34-2012) and by the Animal Experimentation Ethical Committee Darwin (Reference: 02224.02).

Copyright

© 2019, Riva et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 2,700
    views
  • 399
    downloads
  • 34
    citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Martina Riva
  2. Ioana Genescu
  3. Chloé Habermacher
  4. David Orduz
  5. Fanny Ledonne
  6. Filippo M Rijli
  7. Guillermina López-Bendito
  8. Eva Coppola
  9. Sonia Garel
  10. Maria Cecilia Angulo
  11. Alessandra Pierani
(2019)
Activity-dependent death of transient Cajal-Retzius neurons is required for functional cortical wiring
eLife 8:e50503.
https://doi.org/10.7554/eLife.50503

Share this article

https://doi.org/10.7554/eLife.50503

Categories and tags

Research organism

Further reading

    1. Developmental Biology

    Distinct functions of three Wnt proteins control mirror-symmetric organogenesis in the C. elegans gonad

    Shuhei So, Masayo Asakawa, Hitoshi Sawa
    Research Article Updated

    Organogenesis requires the proper production of diverse cell types and their positioning/migration. However, the coordination of these processes during development remains poorly understood. The gonad in C. elegans exhibits a mirror-symmetric structure guided by the migration of distal tip cells (DTCs), which result from asymmetric divisions of somatic gonadal precursors (SGPs; Z1 and Z4). We found that the polarity of Z1 and Z4, which possess mirror-symmetric orientation, is controlled by the redundant functions of the LIN-17/Frizzled receptor and three Wnt proteins (CWN-1, CWN-2, and EGL-20) with distinct functions. In lin-17 mutants, CWN-2 promotes normal polarity in both Z1 and Z4, while CWN-1 promotes reverse and normal polarity in Z1 and Z4, respectively. In contrast, EGL-20 inhibits the polarization of both Z1 and Z4. In lin-17 egl-20 cwn-2 triple mutants with a polarity reversal of Z1, DTCs from Z1 frequently miss-migrate to the posterior side. Our further analysis demonstrates that the mis-positioning of DTCs in the gonad due to the polarity reversal of Z1 leads to mis-migration. Similar mis-migration was also observed in cki-1(RNAi) animals producing ectopic DTCs. These results highlight the role of Wnt signaling in coordinating the production and migration of DTCs to establish a mirror-symmetric organ.

    1. Developmental Biology

    Spinal neural tube formation and tail development in human embryos

    Chloe Santos, Abigail R Marshall ... Andrew J Copp
    Research Article

    Primary and secondary neurulation – processes that form the spinal cord – are incompletely understood in humans, largely due to the challenge of accessing neurulation-stage embryos (3–7 weeks post-conception). Here, we describe findings from 108 human embryos, spanning Carnegie stages (CS) 10–18. Primary neurulation is completed at the posterior neuropore with neural plate bending that is similar, but not identical, to the mouse. Secondary neurulation proceeds from CS13 with formation of a single lumen as in mouse, not coalescence of multiple lumens as in chick. There is no evidence of a ‘transition zone’ from primary to secondary neurulation. Secondary neural tube ‘splitting’ occurs in 60% of proximal human tail regions. A somite is formed every 7 hr in human, compared with 2 hr in mice and a 5 hr ‘segmentation clock’ in human organoids. Termination of axial elongation occurs after down-regulation of WNT3A and FGF8 in the CS15 embryonic tailbud, with a ‘burst’ of apoptosis that may remove neuro-mesodermal progenitors. Hence, the main differences between human and mouse/rat spinal neurulation relate to timing. Investigators are now attempting to recapitulate neurulation events in stem cell-derived organoids, and our results provide ‘normative data’ for interpretation of such research findings.

    1. Developmental Biology
    2. Genetics and Genomics

    Determining the effects of paternal obesity on sperm chromatin at histone H3 lysine 4 tri-methylation in relation to the placental transcriptome and cellular composition

    Anne-Sophie Pepin, Patrycja A Jazwiec ... Sarah Kimmins
    Research Article

    Paternal obesity has been implicated in adult-onset metabolic disease in offspring. However, the molecular mechanisms driving these paternal effects and the developmental processes involved remain poorly understood. One underexplored possibility is the role of paternally-induced effects on placenta development and function. To address this, we investigated paternal high-fat diet-induced obesity in relation to sperm histone H3 lysine 4 tri-methylation signatures, the placenta transcriptome and cellular composition. C57BL6/J male mice were fed either a control or high-fat diet for 10 weeks beginning at 6 weeks of age. Males were timed-mated with control-fed C57BL6/J females to generate pregnancies, followed by collection of sperm, and placentas at embryonic day (E)14.5. Chromatin immunoprecipitation targeting histone H3 lysine 4 tri-methylation (H3K4me3) followed by sequencing (ChIP-seq) was performed on sperm to define obesity-associated changes in enrichment. Paternal obesity corresponded with altered sperm H3K4me3 at promoters of genes involved in metabolism and development. Notably, sperm altered H3K4me3 was also localized at placental enhancers. Bulk RNA-sequencing on placentas revealed paternal obesity-associated sex-specific changes in expression of genes involved in hypoxic processes such as angiogenesis, nutrient transport, and imprinted genes, with a subset of deregulated genes showing changes in H3K4me3 in sperm at corresponding promoters. Paternal obesity was also linked to impaired placenta development; specifically, a deconvolution analysis revealed altered trophoblast cell lineage specification. These findings implicate paternal obesity-effects on placenta development and function as one potential developmental route to offspring metabolic disease.