Increased glutamatergic synaptic transmission during development in layer II/III mouse motor cortex pyramidal neurons

Abstract

Postnatal maturation of the motor cortex is vital to developing a variety of functions, including the capacity for motor learning. The first postnatal weeks involve many neuronal and synaptic changes, which differ by region and layer, likely due to different functions and needs during development. Motor cortex layer II/III is critical to receiving and integrating inputs from somatosensory cortex and generating attentional signals that are important in motor learning and planning. Here, we examined the neuronal and synaptic changes occurring in layer II/III pyramidal neurons of the mouse motor cortex from the neonatal (postnatal day 10) to young adult (postnatal day 30) period, using a combination of electrophysiology and biochemical measures of glutamatergic receptor subunits. There are several changes between p10 and p30 in these neurons, including increased dendritic branching, neuronal excitability, glutamatergic synapse number and synaptic transmission. These changes are critical to ongoing plasticity and capacity for motor learning during development. Understanding these changes will help inform future studies examining the impact of early-life injury and experiences on motor learning and development capacity.

Introduction

The motor cortex is critical for various functions, ranging from motor control to motor skill learning (Neafsey et al. 1986; Frost et al. 2000; Hooks et al. 2013; Molnár et al. 2014; Ebbesen and Brecht 2017; Lee et al. 2022). Changes in motor cortex neurons, such as dendritic branching and synapse composition, throughout development play a role in plasticity and the capacity for motor learning (Xu et al. 2009; Yang et al. 2009; Yu and Zuo 2011; Papale and Hooks 2018). During the first postnatal weeks in the rodent, the brain matures in structure, circuitry, and synaptic composition. However, specific functional and anatomic changes in the motor cortex principal neurons have not been described rigorously during this period. We examined layer II/III motor cortex principal neuron maturation from the neuronal to the synaptic level using electrophysiology and biochemical measures of glutamatergic receptor subunits.

Motor cortex maturation and plasticity in the first several postnatal weeks are crucial to the capacity to develop fine motor skills and motor learning that occurs throughout infancy and childhood (Martin et al. 2005). Refinement of motor cortex circuitry and an increase in intrinsic excitability have been described and hypothesized to be involved in the ability to learn more complex motor movements over early development (Biane et al. 2015). Several studies have identified a critical shift during the second postnatal week involving changes in cortical circuitry that result in the ability of the motor cortex to process sensory input (Anastasiades and Butt 2012). During this period in rats, the primary motor cortex develops the capacity to process awareness of the sensory consequences of movements through several developmental changes in intracortical and local connectivity (Dooley and Blumberg 2018). During this time, the canonical circuit, which involves the reciprocal interconnectivity of superficial and deep layers, is strengthening in the motor cortex (Capone et al. 2016). Layer II/III motor cortex neurons receive inputs from somatosensory cortex and are crucial to novel motor skill learning (Withers and Greenough 1989; Pavlides et al. 1993; Kleim et al. 1996; Aronoff et al. 2010). These superficial layers of the motor cortex generate attentional signals and aid in motor planning for layer V to generate motor output through the corticospinal tract (Heinzle et al. 2007; Atencio et al. 2009; Capone et al. 2016). During motor learning, layer II/III undergoes changes in dendritic arborization and primary somatosensory inputs to these layers exhibit synapse-specific plasticity (Peters et al. 2014; Papale and Hooks 2018). However, the specific changes in synaptic composition and function in motor cortex layer II/III pyramidal neurons during development have not been described rigorously. These changes likely play an important role in the capacity for motor learning and control that emerges over this period of development.

Motor deficits following neonatal brain injury are common, with a broad spectrum of outcomes, ranging from cerebral palsy to deficits in motor planning, learning and coordination. Early-life injury and seizures alter calcium-permeable AMPA receptor function and plasticity in the CA1 region of the hippocampus (Lippman-Bell et al. 2016). In order to further understand changes that may occur in as a result of neonatal brain injury in motor cortex, this study aims to describe normal synaptic development during this time period.

This study will examine both anatomy and function of motor cortex layer II/III principal neurons through development in the mouse by combining electrophysiology and biochemical measures of glutamatergic receptor subunits. Further knowledge in this area is important to understand how early-life injury may disrupt motor cortex development and lead to motor deficits later in life.

Materials and methods

Animals

All studies were performed in accordance with protocols approved by the Animal Care and Use Committee of the University of Virginia. Unless otherwise stated, all chemicals were obtained from Sigma (St. Louis, MO, USA) or Fischer Scientific (Hampton, NH). Young adult (p30) and neonatal (p10) c57/Bl6 mice of both sexes were used in the current study. A subset of p60 mice of both sexes were used for a limited number of electrophysiology experiments examining ESPCs.

Electrophysiology

Acute slice preparation

Animals were deeply anesthetized using Halothane and then decapitated. Brains were quickly removed and immersed in oxygenated (95% O₂/5%CO₂) ice-cold (0–4°C) slicing buffer (in mM, 65.5 NaCl, 2 KCl, 5 MgSO₄, 25 NaHCO₃, 1.1 KH2PO₄, 1 CaCl₂, 10 glucose, and 113 sucrose; for pups, 300 mOsm, pH 7.35 to 7.45). Obliquely cut (approximately 60°) slices from the frontal lobe (300-350 μm thick) were prepared using a Vibratome (Leica VT1200S, Germany) to obtain the primary and secondary motor cortex. The slices were then transferred to an incubation chamber, where they were kept submerged at 33°C in artificial CSF (aCSF, containing (in mM), 127 NaCl, 2 KCl, 1.5 MgSO₄, 25.7 NaHCO₃, 10 glucose, and 1.5 CaCl₂; 300 mOsm, bubbled with 95% O₂/5%CO₂, pH 7.35 to 7.45).

Patch-clamp recordings

Motor cortex (primary and secondary) layer II/III principal neurons were identified under the differential interface contrast (DIC) microscope and recorded in pups and adults. During the recording, slices were transferred to a recording chamber perfused with aCSF at 2–3 mL/min. Recordings were performed by using a Multiclamp 700A (Molecular Devices, Union City, CA) amplifier. The data were collected using pClamp 10.2 software (Molecular Devices), filtered at 2 kHz, sampled at 10 kHz (Digidata 1440A; Molecular Devices), and stored on a Pentium-based computer. Recording pipettes were fabricated from borosilicate glass (1.5 mm outer diameter, 0.86 mm inner diameter; Sutter Instrument, Novato, CA) using a Flaming-Brown micropipette puller (P-1,000; Sutter Instruments, Novato, CA). EPSCs were recorded in voltage-clamp mode. Tight-seal whole-cell recordings were obtained using standard techniques (Hamill et al. 1981). Recording pipettes had open-tip resistances of 6–8 MΩ and were filled with the internal solution containing (in mM): D-gluconic acid 110, CsOH 110, CsCl 10, EGTA 1, CaCl₂ 1, HEPES 10, Mg-ATP 5 and lidocaine 5, pH 7.3; 290 mOsm. Electrode capacitance was compensated electronically and access resistance was monitored continuously. The recording was terminated if the series resistance increased by 20%. Layer II/III cells were identified visually, patched in voltage-clamp configuration, and recorded in a holding potential of −65 mV for EPSCs. Current-clamp recordings were performed to measure cell membrane properties and fill biocytin into the cell. The recordings were performed in tight-seal (seal ≥1GigΩ) current-clamp mode. The internal solution consisted of (in mM): 135 K-gluconate, 7 KCl, 10 HEPES, 0.5 EGTA, 2.5 NaCl, 4 Mg-ATP, and 0.3 Na-GTP. For biocytin labeling, 5% biocytin was added to the internal solution and continued recording for 20 to 50 minutes.

We evaluated the presence of a GluA2-lacking subunit in the AMPARs by applying its open-channel blocker IEM-1460 (100 μM). We recorded sEPSCs from layer II/III principal neurons in the motor cortex for p10 and p30 animals. After a 5-minute baseline recording, we applied the drug for 10 min.

Immunohistochemistry

Immunohistochemistry was performed on p10 and p30 brains on free-floating 40 μm coronal sections (Sun et al. 2004) after transcardiac perfusion and fixation with paraformaldehyde. Negative controls were performed, staining patterns were reproducible on separate samples, and results were consistent with previous studies using these antibodies (Bordeaux et al. 2010). The antibodies used include: GluA1 subunit (clone RH95, 1:1,000, Millipore), anti-GluA2 (1:500, Millipore), synaptophysin (1:2,000, Abcam), vGluT1 (1:500, Abcam). Sections were imaged using a Nikon Eclipse TI-U, at 20x/0.75 NA, 1,024 × 1,024 frame size. Nine images per z-stack with 5 μm z-distance between each image were stitched with the use of NIS-Elements software. For biotin-filled cells, a 40×/1.30NA lens was used.

Western blots

For western blots, p10 and p30 mice were anesthetized with isoflurane, decapitated and the brains removed. The motor cortex was isolated using microdissection (from Bregma 2.57 mm–0.25 mm) (Paxinos and Franklin 2008) and used for total protein expression. Samples were prepared and analyzed as previously described (Joshi et al. 2017). Antibodies against the GluA1 subunit (clone RH95, 1:1,000, Millipore), GluA2 subunit (1:500, Millipore), synaptophysin (1:10,000, Abcam), vGluT1 (1:500, Abcam) were used along with anti-ꞵ-actin antibody (1:5,000, Sigma-Aldrich). The protein content of the tissue lysates was estimated using BioRad DC™ protein assay as per the manufacturer’s instructions. Standard SDS-PAGE and western blotting techniques were used. The membranes were imaged using a chemiluminescent reagent (Perkin-Elmer). The blots were exposed on a Chemidoc Touch imaging system (BioRad) and then the optical density (OD) of the signal was analyzed using Image Lab software (BioRad). The total expression was calculated as a ratio of OD of the antibody to the OD of β-actin.

Intracellular protein expression

To examine intracellular proteins, a BS3 reagent was used to eliminate surface proteins from detection by cross-linking reaction (Joshi et al. 2017). After isoflurane inhalation, mice were decapitated, brains removed, stored in cryoprotective solution (PBS with 30% sucrose), and sliced. 300 μm slices from the motor cortex area were incubated in ice-cold artificial cerebrospinal fluid (aCSF) with BS3 (1 mg/ml) at 4°C for 40 min with constant shaking. The tissue was then washed twice in ice-cold CSF containing 10 mM Tris-Cl (pH 7.4) to stop the reaction and the motor cortex area was then microdissected. To obtain enough material to use in the western blot assay, slices from two animals were pooled and used as a single replicate.

Biocytin filling and dendrite analysis

One cell per 300-μm slice from layer II/III of the motor cortex was filled with biocytin for 60 minutes and fixed with PFA for 48 h at 4°C, then in PBS for no longer than 5 days, to examine the dendritic arborization. Sections were rinsed three times with freshly made 0.01 M PBS for 10 min each, then incubated in 2% Triton X-100 for 30 minutes. Then, slices were washed three times with freshly made 0.1 M PBS for 10 min each. Slices were incubated in a blocking solution: 10% NGS and 0.3% Triton-100 for 1 h, then washed three times for 10 min each with 0.1 M PBS and stained with streptavidin-AlexaFluor 488 (1:1,000) to examine morphology in 0.3% Triton, 3% NGS at 4°C overnight in the dark. The next day, after 1 h at room temperature, slices were washed three times (10 min each) with 0.1 M PBS. Slices were mounted in Moviol and imaged at 40× magnification on a Nikon Eclipse TI-U microscope. Dendrites were delineated, measured, and analyzed per previously described protocol in Fiji ImageJ (Wang et al. 2019).

Data analysis

Electrophysiology: Amplitude and frequency of sEPSC were analyzed by MiniAnalysis software. Decay kinetics were evaluated using weighted tau (Eyre et al. 2012)(Eyre et al. 2012). Current clamp data were analyzed using Clampex 10.5 and MiniAnalysis. Student paired t-test was used for cells recorded in the same slice (before and after IEM application). As noted in the results, the unpaired t-test, one-way ANOVA, and two-way ANOVA were also used when appropriate. P-values < 0.05 were considered significant.

Immunohistochemistry/western blots: Student unpaired t-tests using GraphPad Prism were used for western blot protein expression analysis and dendritic branching and length analysis. Data are presented as the mean ± SEM, and values P < 0.05 were considered significant.

Results

Motor cortex dendritic branching increases over development

Layer II/III neurons in the motor cortex (primary and second) were identified under the differential interface contrast (DIC) microscope and recorded in p10 and p30 (Fig. 1A and B). The location of the cell and its cell type were assessed post hoc. We observed increased dendritic branching in neurons from p30 animals compared to p10. Quantitively, biocytin-filled neurons from p30 animals had more dendritic branching than those from p10 animals (p10 = 29.6 ± 5.87 vs. p30 = 49.2 ± 5.704, P = 0.044) (Fig. 1C). However, the dendritic branch length did not change (p10 = 51.76 ± 7 vs. p30 = 62.71 ± 4.252, P = 0.218) (unpaired t-test; n = 5 neurons from 4 p10 mice, n = 5 neurons from 4 p30 mice).

We tested resting membrane potential, membrane resistance, and membrane time constant to examine the passive cell properties in pups and adults (n = 10 neurons from 5 p10 mice, n = 8 neurons from 4 p30 mice; unpaired t-test). Resting membrane potential (p10 = −63.54 ± 1.815 mV vs. p30 = −63.05 ± 2.526 mV, P = 0.873) (Fig. 1D), input resistance (p10 = 280.8 ± 13.14 MΩ vs. p30 = 260.0 ± 23.09 MΩ, P = 0.426) (Fig. 1E), and membrane time constant (p10 = 22.33 ± 0.9 ms vs. p30 = 19.78 ± 3.02 ms, P = 0.386) (Fig. 1F) were similar in p10 and p30 neurons.

Neuronal excitability increases with age

We tested active cellular properties of pyramidal neurons from p10 and p30 mice to examine changes in neuronal excitability during development. A series of current pulses of increasing magnitudes (range:-100pA to 180pA, 20pA for each step) were injected to depolarize the neurons to suprathreshold potentials. Examples of firing in response to current injection for p10 and p30 neurons are shown in Fig. 2A and B. The action potential (AP) threshold of p30 neurons was lower than that of p10 neurons (p10 = −32.87 ± 1.31 mV vs. p30 = −48.45 ± 0.91 mV, P < 0.0001, unpaired t-test, n = 10 p10 mice, 8 p30 mice) (Fig. 2C), suggesting that p30 neurons are more excitable. To confirm this, we analyzed rheobase current, which was also lower in p30 neurons (p10 = 111.1 ± 7.48pA vs. p30 = 59.33 ± 11.36pA, P = 0.0012, unpaired t-test) (Fig. 2D). The AP amplitude (p10 = 95.79 ± 2.54pA vs. p30 = 85.18 ± 3.38pA, P = 0.272, unpaired t-test) and width (p10 = 2.47 ± 0.11 ms vs. p30 = 2.50 ± 0.54 ms, P = 0.903, unpaired t-test) was similar between both groups (Fig. 2E and F). A larger after-depolarization potential (ADP) was observed in p30 neurons (p10 = 12.08 ± 0.667pA vs. p30 = 14.8 ± 0.666pA, P = 0.012, unpaired t-test) (Fig. 2G).

We then compared intrinsic firing patterns of neurons in response to depolarization-induced changes in AP frequency. The action potentials were more frequent in response to current injection in p30 neurons compared to p10 neurons (P < 0.0001, two-way ANOVA, n = 10 p10 mice, 8 p30 mice) (Fig. 2H). The first ten AP’s occurred rapidly (at higher frequency) in p30 neurons compared to p10 neurons (P = 0.047) (Fig. 2I). Neuronal excitability increases over the first several weeks of postnatal development.

Glutamatergic synapses and transmission increase with age

Given the overall increase in dendritic branching and excitability from p10 to p30, we next examined development at the synaptic level. Levels of synaptophysin (a general presynaptic marker) expression were significantly higher in p30 compared to p10 (P = 0.001, unpaired t-test, n = 4 samples from 8 p10 mice and 4 samples from 8 p30 mice (Fig. 3A and B), indicating an increased number of synapses.

To examine glutamatergic synapses in the p10 and p30 motor cortex, we performed immunochemical studies of pre- and postsynaptic markers, including vGlut, GluA1, and GluA2. Presynaptic marker vGluT1, involved in packaging glutamate into synaptic vesicles, expression was significantly higher in the p30 mice compared to p10 (P = 0.006) (Fig. 3C and D). We also examined the expression of AMPAR subunits, GluA1 and GluA2, in the p10 and p30 mice. GluA1 subunit expression was significantly higher in p10 than p30 (P = 0.01) (Fig. 3E and F). Conversely, GluA2 subunits expression was significantly higher in p30 compared to p10 (P = 0.025) (Fig. 3G and H).

Given these findings in total expression in GluA1 and GluA2, we then decided to examine functional (surface) expression using BS3 intracellular protein analysis. We found that GluA1 subunit intracellular expression was unchanged in p10 vs. p30 (P = 0.822; n for p10 = 6 samples from 10 mice, p30 = 4 samples from 8 mice) (Fig. 3J). The decrease in the total expression of GluA1 from p10 to p30 and unchanged intracellular expression suggests that there may be more function (surface expressed) GluA1 subunit-containing AMPA receptors in p10 compared to p30. When we examined intracellular GluA2 expression, we found that it was increased in p30 compared to p10 (P = 0.034; n for p10 = 5 samples from 10 mice, p30 = 6 samples from 10 mice) (Fig. 3K). Similarly, the total expression of GluA2 subunit increased in p30, suggesting that the surface expression of GluA2 subunit remained stable during development. The relatively higher surface expression of GluA1 subunit-containing AMPA receptors in p10 mice raises the possibility that during the neonatal period calcium-permeable (gluA1 containing) AMPA receptors were expressed and decline into young adulthood.

Given the increase in glutamatergic synaptic markers over development, we next investigated AMPA-mediated synaptic transmission in pyramidal cells of layers II/III of the motor cortex at p10 and p30. AMPA receptor-mediated spontaneous excitatory postsynaptic currents (sEPSCs) were recorded in p10 and p30 when GABA transmission was blocked by picrotoxin (200 μm) and NMDA current was blocked by APV (50 μm) (Fig. 4A and B). The frequency of sEPSCs in p10 neurons was lower than that obtained at p30 (p10 = 0.33 ± 0.06 Hz vs. p30 = 0.83 ± 0.18 Hz, P = 0.015), but the amplitude was similar (p10 = 18.25 ± 0.55 mV vs. p30 = 18.72 ± 1.08 mV, P = 0.739)(Fig. 4C and D). sEPSC decay time,(p10 = 9.93 ± 1.29 ms vs. p30 = 6.89 ± 0.85 ms, P = 0.078) and rise time (p10 = 1.28 ± 0.08 ms vs. p30 = 1.50 ± 0.13 ms, P = 0.153) were similar in the two groups (Fig. 4E and F) (n = 8 neurons from 4 p10 mice; n = 7 neurons from 4 p30 mice, unpaired t-test). Prior studies in motor cortex layer V pyramidal neurons have suggested that passive and active membrane properties do not significantly change beyond p30 (Perez-García et al. 2021). However, given the possibility that puberty, occurring around p35–50 (Falconer 1984), has some impact on these properties, we performed EPSCs in p60 mice of both sexes. EPSC amplitude (P = 0.929) and frequency (P = 0.235) were not significantly different between p30 and p60 groups (p30 n = 7 neurons from 4 mice, p60 n = 10 from 5 mice; unpaired t-test) (Fig. S1).

The pattern of changes in the GluA1 and GluA2 subunit expression was distinct during development. A relatively lower expression of GluA2 subunits in p10 animals raised the possibility of the expression of GluA2 subunit-lacking AMPARs, which would have a higher calcium-permeability, in the neonatal mice. We evaluated the expression of functional GluA2 subunit-lacking AMPARs using IEM1460, a selective open-channel blocker of GluA2-lacking AMPARs, and recorded sEPSCs. In p10 neurons, IEM 1460 reduced the amplitude of sEPSCs (P = 0.03) (Fig. 4G and H) but not the frequency (P = 0.871) (Fig. 4I). IEM-1460 did not affect the sEPSCs recorded from p30 neurons (Fig. 4J and K). These findings indicate higher expression of GluA2 subunit-lacking AMPARs in layer II/III neurons in p10 animals, making them more Ca²⁺ permeable. This is supported by our findings in the intracellular expression analysis of GluA1 and GluA2 subunits in western blot experiments above.

Discussion

We examined the development of pyramidal neurons in layer II/III of the mouse motor cortex from the neonatal period to early adulthood using a combination of electrophysiology and biochemical measures of glutamatergic receptor subunits. We found that layer II/III motor cortex principal neurons increase dendritic branching and excitability from the neonatal to young adult age at the neuronal level. We demonstrate that synapse number and glutamatergic synaptic transmission increase over the first several postnatal weeks. Concurrently, glutamatergic synapses increase with a higher GluA1:GluA2 subunit ratio in the p10 mice, rendering increased calcium permeability in younger mice.

Maturation of synaptic composition and function over the first several postnatal weeks are critical to gaining fine motor and motor learning skills throughout early development (Xu et al. 2009; Yang et al. 2009; Yu and Zuo 2011; Papale and Hooks 2018). Changes in excitability and circuitry in the motor cortex have been described previously and theorized to be central in gaining the capacity for complex motor learning (Biane et al. 2015). In the rodent, the neonatal and infant periods involve a shift in cortical circuitry that results in the motor cortex processing and integrating sensory input (Anastasiades and Butt 2012; Capone et al. 2016; Dooley and Blumberg 2018). Layer II/III pyramidal neurons, which we have focused on in this study, receive inputs from somatosensory cortex and integrate them into motor planning and attention to generate motor output via layer V pyramidal neurons to the corticospinal tract (Withers and Greenough 1989; Pavlides et al. 1993; Kleim et al. 1996; Capone et al. 2016;Heindorf et al. 2018 ; Benedetti et al. 2020). Significant changes occur in motor cortex layer II/III during motor learning, including increased dendritic arborization and synapse-specific plasticity (Xu et al. 2009; Yang et al. 2009; Yu and Zuo 2011; Ma et al. 2016; Papale and Hooks 2018). In addition, motor learning is associated with increased surface expression of GluA1-containing AMPAR on dendritic spines in motor cortex layers II/III and V (Kida et al. 2016; Roth et al. 2020; Miyamoto et al. 2021).

Several studies have demonstrated that postnatal neuronal and synaptic development likely differs by region and layer, given the diverse functions of each of these regions (Petit et al. 1988; Núñez-Abades and Cameron 1994; Zhu 2000; Zhang 2003; Carrascal et al. 2010; Perez-García et al. 2021). At the neuronal level, passive cell properties in layer V of the rat motor cortex are nearly established by week three of postnatal development, with the most dramatic changes happening during weeks 1–2 (Zhang 2003; Perez-García et al. 2021). At the synaptic level, several previous studies have examined changes in synaptic function, receptor composition, and numbers through development, but the majority of these studies have focused on the hippocampus or the entire cortex (Monyer et al. 1991; Pellegrini-Giampietro et al. 1991; Burnashev et al. 1992; Durand and Zukin 1993; Arai et al. 1997; Martin et al. 1998). Glutamate receptor subunit, GluA1, expression in the developing rodent hippocampus increases from the neonatal period to young adulthood (Durand and Zukin 1993; Sans et al. 1996; Arai et al. 1997; Biane et al. 2015; Tzakis and Holahan 2020) and declines with further aging. However, data on GluA1 expression in the rodent cortex are less clear; GluA1 expression in the entire cerebral cortex peaks around p10–14, then declines until adulthood (Monyer et al. 1991; Burnashev et al. 1992; Arai et al. 1997). Specifically, in the frontal cortex, GluA1 mRNA expression was shown to peak around p28, then decline into adulthood (Durand and Zukin 1993). Similarly, expression of the GluA2 subunit in the hippocampus increases over the course of development from neonate to adult (Durand and Zukin 1993; Sans et al. 1996; Tzakis and Holahan 2020) and in some regions of the cortex, such as the frontal cortex (Pellegrini-Giampietro et al. 1991; Durand and Zukin 1993). Previous studies did not focus specifically on layer II/III principal neurons of the motor cortex, which is crucial to integrating sensory input and provides excitatory drive to the pyramidal tract neurons (Heindorf et al. 2018; Benedetti et al. 2020). Similar to our findings, corticospinal neurons exhibit increased intrinsic excitability during postnatal development (Biane et al. 2015). However, contrary to our results in layer II/III of the motor cortex, layer V motor cortex neurons became less excitable over the first several postnatal weeks (Benedetti et al. 2020; Perez-García et al. 2021). Specific changes in layer II/III of motor cortex may differ from other layers and regions, as this region must maintain plasticity for motor learning that continues into adulthood (Peters et al. 2014; Papale and Hooks 2018).

Here, we have limited these studies to p10 and p30 timepoints because they are of great importance in future studies of the impact of neonatal brain injury on motor cortex development and resultant motor deficits. The most common neonatal brain injury is hypoxic–ischemic injury in the near and full-term gestation neonate and commonly leads to motor deficits (such as cerebral palsy and motor coordination disorder). The postnatal day 10 mouse model of hypoxia-ischemia is a widely used animal model of near and full-term brain injury (Rice, et al. 1981; Yager and Ashwal 2009; Albertsson et al. 2014; Hamdy et al. 2020). We hypothesize that injury during the second postnatal week, a critical time in motor cortex development, will lead to abnormal synaptic development and contribute to motor deficits. Postnatal day 30 in the mouse neurologically translates roughly to the young adult period. Previous studies have found in layer V motor cortex pyramidal neurons, passive and active cell properties change very little beyond p30 (Perez-García et al. 2021). Our limited data comparing EPSCs in p30 and p60 layer II/III motor cortex principal neurons supports this finding. Clinically, the majority of motor deficits resulting from neonatal brain injury are evident prior to young adulthood. The findings we present here will allow ongoing work to examine the specific changes in motor cortex maturation resulting from injury during the neonatal period.

Using a combination of electrophysiology and biochemical measures of glutamatergic receptor subunits, we have demonstrated that layer II/III principal neurons of the mouse motor cortex undergo several important changes over the span from neonate to young adult. There is an increase in dendritic branching and excitability at the neuronal level. During this critical period, the concurrent higher expression of calcium-permeable AMPA receptors in p10 mice may play an important role in dendrite growth and synapse formation. Neuronal and synaptic maturation in layer II/III of the motor cortex from the neonatal period to young adulthood plays an important role in plasticity and the capacity for motor learning during development. Future studies will focus on how early-life injury during this critical window disrupts normal synaptic development in the motor cortex and the role in motor deficits following neonatal brain injury.

Acknowledgments

Thanks go to the University of Virginia Keck Center for Cellular Imaging for the use of the Leica tauSTED super-resolution confocal microscope.

Funding

This work was supported by the National Institutes of Health, NINDS K08NS101122 (to J.B.), R01NS110863 (to S.J.), R01NS120945 (to J.K.), R37NS119012 (to J.K.), R01NS040337 (to J.K.), the UVA Brain Institute and UVA Department of Pediatrics.

Conflict of interest statement: None declared.

References