Somato-dendritic decoupling as a novel mechanism for protracted cortical maturation
- 2k Downloads
Both human and animal data indicate that disruption of the endogenously slow maturation of temporal association cortical (TeA) networks is associated with abnormal higher order cognitive development. However, the neuronal mechanisms underlying the endogenous maturation delay of the TeA are poorly understood.
Here we report a novel form of developmental plasticity that is present in the TeA. It was found that deep layer TeA neurons, but not hippocampal or primary visual neurons, exist in a protracted ’embryonic-like’ state through a mechanism involving reduced somato-dendritic communication and a non-excitable somatic membrane. This mechanism of neural inactivity is present in intact tissue and shows a remarkable transition into an active somato-dendritically coupled state. The quantity of decoupled cells diminishes in a protracted and age-dependent manner, continuing into adolescence.
Based on our data, we propose a model of neural plasticity through which protracted compartmentalization and decoupling in somato-dendritic signalling plays a key role in controlling how excitable neurons are incorporated into recurrent cortical networks independent of neurogenesis.
KeywordsPyramidal Neuron Neocortex Excitability Remote Memory
Delayed functional maturation of the temporal association cortex (TeA) during early postnatal development is a well-recognized phenomenon of brain maturation that is conserved across species [1, 2, 3, 4, 5]. Cognitive neuroscientists have long realized that it is during this critical window that large-scale developmental changes in the brain co-occur with marked changes and expansion in cognitive capabilities that seem to follow a defined hierarchical chronological pattern [6, 7, 8, 9, 10]. This pattern of development is thought to allow more specialized higher order information processing streams to incrementally emerge, but only after lower level neural afferents from primary cortices become more mature and stable [9, 11, 12, 13].
Largely derived from early maturing brain circuits, popular neurobiological models of brain development emphasize that the intrinsic neural processing capabilities of cortical networks develop synchronously [8, 14] and are present very early during development to facilitate activity-dependent refinement of circuits through selective elimination of initial exuberant growth [7, 15, 16]. However, a major challenge for this type of developmental model is that it is difficult to reconcile with the findings of different, and often opposing, trends of protracted maturation trajectories, both between cortical regions and within high-order recurrent cortical networks (e.g. TeA versus prefrontal cortex) [13, 17, 18]. Furthermore, during the postnatal period, cognitive functional development is often associated with a protracted increase in activation areas and responses along the more specialized information processing streams in the TeA [2, 3, 5, 10, 19]. This supports both theoretical and empirical findings that limiting initial computational resources may actually facilitate, rather than hinder, normal cognitive development in recurrent neuronal networks [6, 7]. These seemingly contradictory findings, therefore, raise questions regarding the possible neurobiological mechanism of protracted high-order cortical maturation.
The neuronal mechanisms underlying delayed maturation of the TeA are poorly understood. Previous models of early postnatal maturation have often focused on structural constraints such as synaptic and/or dendritic complexity [7, 8, 15]. However, there is accumulating evidence that developmentally regulated functional constraints may also be important. For instance, electrophysiological recordings and functional imaging have shown that the TeA lacks significant excitability and functional connectivity during the first month of postnatal life [1, 20], and accelerating TeA maturation during early postnatal development can disrupt normal behavioural development [20, 21]. These observations are also in line with data consistently reporting that electrical activity is not required for the establishment of basic neuronal morphology [22, 23, 24]. Together, these findings suggest that neurobiological processes that support independently regulated neuronal morphological and electrophysiological maturation may provide an important additional mechanism of protracted functional cortical maturation.
Reduced TeA excitability despite continued growth
Neuronal dormancy is associated with poor somato-dendritic coupling and a specific lipophilic cytosolic structural organization
The fact that dendritic electrophysiological maturation can precede that of the soma  and that dendritic surface area is several-fold greater than that of the soma  indicates that the majority of somatically recorded current reflects dendritic conductances. Indeed, previous experimental evidence has directly confirmed this  and is consistent with the fact that hyperpolarizing-activating cationic current (I h), which is commonly observed in somatic recordings , is not present in the soma but increases in density with increasing distance from the soma . Therefore, one possibility may be that there is a complete and uniform lack of functional conductances in TeA dendrites. While unlikely, we nevertheless tested this by applying the same recording procedure directly to dendrites (Additional file 2: Figure S2). Although TeA dendrites are generally quite delicate and difficult to record from, when dendritic recordings were obtained and in contrast to somatic recordings, they indeed had significantly greater membrane conductance at both hyperpolarized (–100 mV) and, in particular, depolarized (+50 mV) potentials even in the same neuron (Additional file 2: Figure S2). Moreover, all dendritic recordings exhibited outward rectification (Additional file 2: Figure S2) and, in at least 50 % of dendritic recordings, inward spike currents with peak amplitudes ranging from 0.24–2.9 nA (mean = 1.7 ± 0.8 nA; not shown) were also evident with step depolarization. Thus, it does not appear that the dormant phenotype results from a general lack of membrane conductances (also see below).
A second possibility may be that the soma of the dormant phenotype is functionally compartmentalized. As a functionally isolated soma would fail to incorporate the dendritic capacitance load , we first measured capacitance values of dormant and spiking neurons. Indeed, we found that dormant neurons consistently exhibited a significantly smaller membrane capacitance than that of spiking cells (Additional file 3: Figure S3), similar to what was previously observed in immature neurons [30, 31, 32, 33]. In fact, as the recorded membrane capacitance of dormant neurons is too large to represent a cell-attached configuration  and too small to incorporate the dendritic compartment , we calculated the expected somatic membrane capacitance based on dimensions of the soma. As predicted, the calculated somatic values correspond to the experimentally measured capacitance values of the dormant phenotype (see Additional file 3: Figure S3).
Given that a greater degree of intracellular membrane biogenesis is needed during neuronal maturation , we also tested to see if there was a difference in the spectral density of intracellular binding of the lipid-binding dye FM1-43 between spiking and dormant cells. The rationale for this is that it may not only help explain previous reports that complete labelling of the somatic and dendritic compartments with aqueous dye can sometimes be more difficult in neurons from younger animals during intracellular recording [43, 44], but it may also provide insight into a potential mechanism contributing to somato-dendritic decoupling. To this end, the amphipathic dye FM1-43, which does not fluoresce when in an aqueous volume fraction, was delivered intracellularly through the patch pipette during recordings done to verify electrophysiological phenotype. It was found that dormant cells exhibited a significantly higher fluorescence spectral density signal than that of spiking cells (Fig. 3g–h), consistent with an intracellular somatic structure involving a densely ordered cytoplasmic space .
TeA neuronal dormancy decreases with advancing age
Neural circuit decoupling and dormancy rather than neurogenesis
An intriguing possibility related to protracted functional maturation of the TeA network would be to produce new neocortical pyramidal neurons through the process of postnatal neurogenesis. Although the features of dormant neurons are reminiscent of very immature neurons found during adult neurogenesis , an important feature of the postnatal neocortex is that it does not appear to produce new neurons  (Additional file 5: Figure S5). In addition, although we did not specifically examine the chemical nature of dormant neurons, i.e. whether they are destined to an interneuron fate, the high proportion of this phenotype (about 75 %) during early postnatal development is inconsistent with the relatively few (<25 %) interneurons present in the neocortex . Rather, our data indicate that somato-dendritic decoupling has led to a dormant phenotype and a previously under-appreciated maturational stage of pyramidal cells, an assertion supported by their morphology and the fact that these neurons also have sub-cortical projections .
Physiological basis of somato-dendritic decoupling
Polarized neurons are highly compartmentalized and can maintain discrete and independent signalling mechanisms within restricted sub-cellular domains [54, 55, 56]. Many previous reports indicate that the somatic compartment of an immature neuron can possess few cationic conductances, especially when it is artificially isolated from all neurites  or at a developmental stage absent of complex dendritic structure [30, 31, 32, 33]. On the other hand, axons and dendrites isolated from cell somata can still maintain RNA and protein synthesis, as well as membrane trafficking [54, 55, 56]. Indeed, dendritic and axonal growth can proceed at different rates and operate via different mechanisms , and somatic activity does not appear to be required for axonal development . Consistent with these studies, decoupled TeA neurons appear to develop and retain basic dendritic morphology and long axonal projections , suggesting the presence of a unique, but not entirely unexpected, homeostatic state allowing cell maturation to be carried out in a highly parallel and compartmentalized fashion.
The significantly reduced somato-dendritic coupling in dormant cells is supported by previous studies [43, 44] as well as our own observations, indicating that neurons in younger animals can sometimes be more difficult to rapidly and extensively fill with intracellular aqueous (particularly Alexa), but not amphipathic dye. In coupled neurons, however, we have observed mature electrophysiological properties and individual pyramidal neuron firing patterns comparable to those reported in the literature [30, 60, 61, 62]. This suggests that the mechanism underlying somato-dendritic decoupling is likely modifiable under physiological conditions, especially when considering the fact that a whole range of physiological neuronal firing patterns can be produced by a two-compartmental computational model described by two parameters: dendritic size and somato-dendritic interface resistance . In fact, the thin intracellular cytoplasmic space and complex internal protein-lipid structure characteristic of immature neurons may significantly increase interface resistance [42, 45, 64].
Maintenance of somato-dendritic decoupling
Several regulatory mechanisms of neuronal maturation may influence the degree of somato-dendritic decoupling and viability of decoupled cells. For example, in many immature cell types, the early expression of functional calcium channels can give rise to calcium-mediated action potentials that are conserved during transient periods of development before transitioning into a more mature sodium-mediated action potential phenotype [28, 65]. In the rat cerebellum, activation of immature granule cells results in calcium-mediated action potentials during early (but not late) postnatal development , while in insects, the period with which calcium-mediated action potentials operate coincides with early dendritic remodelling of pupal motoneurons . A multitude of developmental genomic and epigenetic programs are hosted in the soma, and their activation often requires external stimulation and calcium influx [16, 28]. The fact that in decoupled TeA neurons calcium- rather than sodium-mediated spikes can be recorded following electrically disrupted decoupling and the fact that sodium-mediated spikes are present in spiking neurons from more mature tissue (Additional file 6: Figure S6) further support this developmental model in the TeA. Decoupling may therefore be an important mechanism in regulating the accessibility of these signals on the somatic genomic/epigenomic machinery involved in cellular maturation.
A second mechanism that may contribute to the prolonged dormancy of TeA neurons may be related to the hypothesis that neuronal maturation in the mammalian cortex may be governed by regional specific cues [68, 69, 70] and/or the hierarchical pattern of afferent and efferent connectivity . In this regard, the maturation speed of TeA neurons may be influenced by several factors, including the levels of afferent activity and trophic signalling. For instance, trophic signalling via brain-derived neurotrophic factor (BDNF) is important in nervous system maturation [71, 72]. It can stimulate dendritic growth  (also see Additional file 3: Figure S3) and induce somato-dendritic communication and transcriptional regulation , and it can up-regulate functional sodium channel expression . Following BDNF signalling, positive feedback between voltage-gated and capacitive calcium current, as well as autocrine and paracrine trophic signalling [76, 77], may therefore allow for somato-dendritic coupling to emerge during postnatal dendritic remodelling and Golgi expansion .
Finally, the absence of significant somatic cationic channels may further serve a role in protecting decoupled cells from cationic channel-dependent apoptotic cell death and elimination . However, the lack of significant somatic potassium conductance in the decoupled state does suggest an atypical ionic basis for the establishment of somatic resting membrane potential in the decoupled state. While chloride conductance and cationic co-transporters are crucial in ion homeostasis , they may also play a role in establishing resting membrane potential under conditions of reduced potassium conductance [80, 81].
Somato-dendritic decoupling and neuronal dormancy represent a unique form of brain plasticity associated with cognitive network maturation in the TeA. Our data support the idea that, independent of neurogenesis, new functional units may be gradually added to the TeA cortical network with increasing proportions of somato-dendritically coupling spiking neurons. Whether a similar type of plasticity also exists in other cortical layers and/or in other association cortices remains unclear. Nevertheless, the polysensory TeA in rodents forms a large-scale recurrent neuronal network and shares many structural and functional similarities to that of primates, including various forms of memory and attention control [53, 82, 83, 84, 85, 86]. Pharmacological agents or pathological conditions that alter the maturational trajectory of the temporal lobe may therefore have a detrimental impact on cognitive development [87, 88]. Indeed, exposure to valproic acid (VPA), which significantly increases the risk of autism in children , not only stimulates TeA dendritic growth and reduces neuronal dormancy in rodents, but it can also lead to pathological features characteristic of autism [20, 21]. These findings, when taken together, indicate that decoupling and neural dormancy may play a very important role in normal cognitive development.
In vitro cultures
All experimental protocols were approved by the University of Calgary Conjoint Faculties Research Ethics Board (AC12-0239). Cells were prepared by a standard protocol. The dimensions and location of the cortical regions of the P0 rat were referenced from the Atlas of the Developing Rat Brain . The entire hippocampus (Hip), primary visual cortex (Oc1) and the polymodal temporal association cortex (TeA), also known as area Te2 which is also part of the visual system [91, 92], were removed by microdissection and incubated in sterile filtered enzymatic solution containing CaCl2 (Sigma, Oakville, ON), L-cysteine (Sigma), EDTA and 20 U/ml papain (Worthington, Lakewood, NJ) at 37 °C for 30 min. All cells were then washed with fresh media and dissociated by triturating though a set of three flame-prepared Pasteur pipettes of decreasing calibre. Cells were counted with a hemocytometer and cultured on glass coverslips (previously coated with poly-D-lysine and laminin) and placed in a 24-well plate at a culture density of about 175 × 103 cells/cm2. Cultures represent cells from multiple (up to 10) pups. The final volume of growth medium was maintained at 800 μl. One half of the media was replaced every 3–4 days. Culture media consisted of Basal Medium Eagle supplemented with glucose, B-27, penicillin-streptomycin, Na pyruvate, HEPES (Sigma), L-glutamine (Sigma), and 4 % fetal bovine serum (FBS). Recordings were obtained at room temperature (RT) from cells immersed in extracellular bath solution (EBS) containing (in millimolar units): 3.0 CaCl2, 2.0 MgCl2, 135 NaCl, 5.0 KCl, 10 glucose, 5.0 HEPES and a pH and osmolality of around 7.3 and 310 mOsm respectively. Cultures were fixed for 20 min in 4 % paraformaldehyde (PFA) at RT and washed with PBS three times for 15 min each. This was followed by a 1-hour incubation in blocking solution containing 0.4 % bovine serum albumin (BSA), 0.1 % Triton X-100 and 0.02 % normal donkey serum (Invitrogen) in PBS. Primary antibodies were incubated with the cultures overnight at 4 °C. After three washing steps in blocking solution for 20 min each, primaries were detected by incubation for 2 hours at RT with fluorescently conjugated secondary antibodies raised in donkey (Chemicon, Temecula, CA). To delineate the plasma membrane of cells, we labelled them with the carbocyanine dye DiI. The best results were obtained when DiI labelling was performed after immunocytochemistry. DIV20 cultures were fixed for 20 min in 4 % PFA at RT and washed with PBS three times for 15 min each. This was followed by 1-hour incubation in blocking solution containing 0.4 % BSA and 0.02 % normal donkey serum (Invitrogen) in PBS. Cultures were incubated for 48 hours at 4 °C in mouse α-NeuN (Chemicon) diluted in blocking solution with 0.001 % Triton X-100 to improve antibody penetration. Higher concentrations of Triton-X resulted in deleterious effects on subsequent DiI labelling. After three washing steps in blocking solution for 20 min each, the primary was detected by incubation for 2 hours at RT with a Cy5 conjugated secondary antibody raised in donkey (Chemicon). NeuN-stained cultures were then co-labelled with one of the following primary antibodies: goat α-NaCH type Iα (Santa Cruz, sc 31451, 1:200) or rabbit α-sodium bicarbonate co-transporter (NBC) (Millipore AB3208, 1:100), which were detected by Alexa488 conjugated secondary of the appropriate species. The nuclear stain DAPI was also included. Cultures were then incubated for 2 hours at 37 °C in a working solution of DiI (about 40 μg/ml) (Molecular Probes). Cultures were washed three to five times in PBS to remove the unbound dye. In each analysis group, cultures were processed for immunocytochemistry and DiI labelling completely in parallel, and images were captured with identical light and camera settings. For analysis of single-cell and spine morphology, cultured neurons were transfected with a pEGFP plasmid using Lipofectamine 2000 (Invitrogen) according to the manufacturer instructions.
Morphological quantification of single GFP transfected neurons
To avoid sampling bias, we measured all GFP transfected neurons with neurites that were clearly visible and not obscured by other cells. When the processes of a neuron extended beyond the field of view, several images were taken and stitched together. Neurons were identified by expression of neuronal nuclear antigen (NeuN) (after DIV5) or basic morphology (before DIV5, due to low expression levels of NeuN at early ages). Neurites were traced with NeuronJ. For spine analysis, images were acquired from DIV20 GFP-expressing neurons at high magnification using an oil immersion 100× objective (Olympus). The number and length of dendritic spines and filopodia were measured from the primary dendrites, proximal to the cell body and extending as long as they remained in focus (for segments longer than 10 μm). We did not correct for spines hidden beneath or above the dendrite; therefore, spine densities likely underestimate the true value. Filopodia were defined as dendritic protrusions that were >3 μm in length. Spines were defined as protrusions <3 μm in length, and similar to those reported previously , we included protrusions that emanated laterally greater than about 0.5 μm from the dendritic shaft. The observer was not blinded as the neuronal morphology of the TeA is visibly different. However, to minimize potential bias , a single observer scored spine densitometry for all regions. Spines were categorized as mushroom, stubby and thin based on the morphological criteria of Peters and Kaiserman-Abramof . The number of dendritic spines of each type were counted using the Cell Counter plug-in in ImageJ. The spine density was calculated by dividing the total number of spines by the length of the dendrite analysed. An average spine density was obtained for each neuron analysed.
Tissue sections were used to evaluate neuronal dormancy under more physiological conditions from a total of 57 animals. This was based on our earlier pilot data indicating that juvenile tissue likely harbours a large proportion of the dormant phenotype, and the large effect size (d = 0.8) that would be expected with a completely functional network in mature tissue (power = 0.8). Slices (250–450 μm) were submerged in a recording chamber in oxygenated (95 % 02; 5 % CO2) artificial cerebrospinal fluid (aCSF). Recordings were targeted to layer V of the posterior sector of the ventral TeA-ectorhinal cortex (also known as area Te2) as this is the major output layer of the TeA. The TeA region is identified as the region located dorsal to the rhinal fissure (e.g. Plates 36–44 in ). Frontal and parahorizontal tissue sections were prepared on a Leica vibrotome (Germany) from Sprague-Dawley and Long-Evans rats. The aCSF had a final pH and osmolality of around 7.4 and 300 mOsm respectively and contained (mM): NaCl, 110; KCl, 3.5; MgCl2, 1.5; NaHCO3, 26; CaCl2, 2; glucose, 10. Recordings were made at 30–33 °C. For single-cell processing, cells were fixed after recording in 4 % PFA overnight at 4 °C. Slices were then washed with PBS three times for 10 min each and incubated in blocking solution (10 % normal goat serum, 0.25 % Triton-X 100, 0.1 M phosphate buffer) for 1 hour. Slices were then incubated in streptavidin-Cy3 (1:1000; Jackson Immunoresearch, West Grove, PA) and mouse monoclonal NeuN antibody (1:500; Chemicon International) in blocking solution for 1 hour at RT. Slices were then washed with PBS three times for 10 min and incubated with secondary fluorescence-conjugated (Alexa488) anti-mouse antibody at RT for 1 hour. Finally, slices were washed in 1× PBS three times for 10 min and then mounted, dehydrated and coverslipped with mounting medium. For 4′,6-diamidino-2-phenylindole (DAPI) staining, 20–40 μm thick sections were cut and stained with DAPI (1:104) for 5 min at RT followed by a PBS rinse three times for 5 min. For mouse-NeuN and rabbit-NF-200, sections were washed three times for 10 min each with PBS and incubated with blocking solution (PBS containing 10 % NGS and 0.1 % Triton X-100) for 1 hour. Sections were then incubated with primary antibody mouse NeuN (1:500) and rabbit NF-200 (1:800) overnight at 4 °C or 1 hour at RT in blocking solution. Sections were then washed three times for 10 min each in PBS and exposed for 1 hour with donkey anti-mouse cy3 (1:250) and goat anti-rabbit cy3 (1:250) at RT diluted in PBS. Sections were then washed three times for 10 min each in PBS and mounted in aqua polymount.
The patch electrode (DC resistance; 6–8 MΩ) solution contained (mM): K-gluconate, 120; KCl, 10; HEPES, 10; Na-GTP, 0.2; Mg-ATP/Na-ATP, 4 or 110 K-gluconate, 10 HEPES, 1.0 EGTA, 20 KCL, 2 MgCl2, 2.0 Na-ATP, 0.25 Na-GTP, 10 di-tris-phosphocreatine, and had a pH and osmolality of around 7.3 and 285 mOsm respectively. The bicarbonate recording solution consisted of 100 mM KCl and 50 mM NaHCO3. The pH was adjusted with HCl and NaOH. Under light microscopy (Zeiss Axioskop Microscope; Carl Zeiss, Germany), neurons were visually selected based on their pyramidal or ovoid morphology  both in vitro and in situ. Bright field images of the same visual field used for immunohistochemical analysis were also captured to confirm that visually identified neurons were positive for neuronal markers (i.e. NeuN or MAP2). Biocytin was also included in the pipette for co-labelling with the neuron-specific marker NeuN in some experiments (Additional file 4: Figure S4).
Obtaining patch-clamp recordings for detailed electrophysiological characterization
The recording configuration was obtained by releasing the positive pipette pressure and simultaneous application of a negative pipette pressure by mouth. Care was taken to minimize mechanical perturbation of the cell/soma by ensuring that only local (i.e. the area under the pipette) membrane deformation and patch rupture occurred. This process could also be confirmed visually under DIC microscopy in some instances. In cases that were not visually confirmed, the amplifier was switched into current-clamp mode to evaluate the temporal dynamics of the recording configuration and the resting voltage potential (I = 0). The former is an important and easy step to rule out the fast membrane charging times associated with a cell-attached configuration despite the ability of still being able to record synaptic and action potential events in cell-attached mode . Additional suction, however, often resulted in intracellular aspiration and cell deformation and/or loss of the patch.
Recordings from neurons in immature cortex typically exhibit more depolarized membrane potentials (V m) (–30 to –60 mV) that can be maintained throughout the recording session . Indeed, in both cell phenotypes observed in the present study these typical membrane potentials were obtained following the patching procedure and could be maintained throughout the recording period. For spiking cells, there was an immediate polarization in V m that remained stable for the duration of the recording that is typical of patch-clamp recordings . The dormant phenotype, on the other hand, exhibited a different pattern; after the initial shift in V m following the patching procedure, the V m started to move toward and stabilized at more hyperpolarized potentials (e.g. –95.4 ± 10.6 mV; n = 10). However, this hyperpolarizing shift in V m was prevented when using a bicarbonate pipette buffer solution (–32.7 ± 6.8 mV, n = 7) that would help prevent a CO2 gradient-dependent acidification at the membrane interfacial region . Indeed, when we acidified the bicarbonate solution to effectively reduce free bicarbonate, we observed a strong effect in the opposite direction to that observed under increased free bicarbonate conditions: significantly decreased conductance (Additional file 7: Figure S7) and further V m polarization (–138.5 ± 6.5 mV; n = 8; F (2,22) = 32.09 p < 0.0001 and all three conditions are significantly different from each other, Tukey’s post hoc test p ≤ 0.01) that moved away from the chloride reversal towards the calculated Na-K-ATPase reversal based on the estimated free energy of ATP hydrolysis .
Immunoreactivity and DiI labelling were examined with an Olympus confocal FV300 laser scanning system and BX50 upright microscope using a 60× UplanApo 1.20 NA water immersion objective. Images were captured in sequential mode in double labelling experiments. Z-stack scans were taken with 1-μm optical sections. Fifteen images were captured from each coverslip by translating the stage in a systematic and unbiased fashion. Optimal light, amplifier and camera settings were defined for each of the fluorochromes used and kept consistent for all images captured. To quantify peri-plasmalemmal channel/transporter expression with confocal microscopy, we analysed the XY plane of the NeuN-positive DiI-labelled cell that was closest to the coverslip (i.e. the first Z plane where the ventral DiI signal becomes evident). We chose to analyse this flat basal part of the cellular membrane because (1) it could be identified objectively, and (2) it provided a larger surface area to analyse than the top or side slices due to their curvature. These regions of interest were defined using ImageJ (NIH, W. Rasband) and their signal intensity was measured.
Transmembrane potentials are based on field strengths required for microtubule translocation . We found that extracellular electrical stimulation on the order of 101–102 V/cm helped promote decoupling (n = 5). Extracellular electric fields can be related to transmembrane potential by a first approximation by the equation ΔV m = –(3/2)E acosθ, where E represents the extracellular electric field strength, a is the cell radius and θ equals the polar angle . This allowed us to localize stimulation to the cell we are recording from to see if we could artificially disrupt the decoupled state. This assumes that (1) the electrically insulated region is roughly spherical, (2) it has a high membrane resistance and (3) it lacks voltage-dependent changes in conductance, all of which are defining characteristics of the decoupled phenotype under standard conditions (see Results). For example, an extracellular electric field of 5 × 101 V/cm, sufficient for microtubule translocation , would be expected to generate a maximum regional transmembrane around –150 to –200 mV under the same recording conditions as extracellular stimulation (ΔV + V soma). Thus, stimulation consisted of sustained repetitive short (e.g. 1 ms) duration hyperpolarizing pulses 1–10 Hz, or several series of longer duration pulses on the order of 101 seconds. Shorter duration stimulation often required even greater intensity stimulation, and the process was facilitated with incremental increases in stimulation intensity following each series of stimulations.
The membrane impermeable amphipathic dye FM1-43 was included in the pipette for intracellular cytosolic structural labelling. To normalize signals from different experiments, FM fluorescence measurements were determined by subtracting the background signal from the mean fluorescence signal in the somatic region. Fluorescence images were analysed using ImageJ (NIH) software. Fluorescence images were captured on-line with a Nikon CCD camera attached to an Axioscope 2 FS-mot Carl Zeiss microscope.
Additional control experiments
There are four clear lines of evidence confirming that the decoupled phenotype is observed under a whole-cell recording configuration. First, the soma can be labelled when different membrane impermeable dyes are included in the pipette (e.g. see Fig. 3). Second, the V m can be changed by using different pipette solutions (see above). Third, pipette dialysis, the hallmark of a whole-cell recording configuration, with hypotonic or hypertonic pipette solution leads to cell shrinkage or cell swelling/dimpling respectively (Additional file 8: Figure S8). Finally, and most notably, despite a lack of recordable cationic current, anionic current can be measured when a bicarbonate pipette solution is used and can be blocked by bath application of anionic channel blockers (Additional file 7: Figure S7). Importantly, this confirms current measurement at the level of the outer plasmalemma.
For the tonicity experiments, strong changes in tonicity are needed to provide the necessary driving force to override the endogenous compensatory processes and allow visualization of a response . Thus, for the hypotonic condition, the standard pipette solution (285 mOsm) was diluted by adding deionized water (e.g. 1.5 ml of deionized water to 2 ml of pipette solution). For the hypertonic condition, approximately 280 mM of magnesium chloride was added to the standard pipette solution. Note that the chloride counterion likely has little effect on intracellular tonicity as it is one of the most prevalent ions involved in endogenous intracellular osmoregulation .
Signal acquisition and analysis was accomplished using the Multi-clamp 700A or Axopatch 200B and DIGIDATA 1322A 16-bit data acquisition system (Molecular Devices). Data was typically low-pass filtered at 2–8 kHz and digitized at ≥10 kHz. Intrinsic properties were determined according to methods previously described [30, 102, 103]. Data are expressed as mean ± SEM, and the statistical test is noted when used. N represents cells unless otherwise specified.
We would like to thank Drs. Bourque, Bains, Pittman, Lukowiak, Horne and Clarke for helpful comments, and Sharmila Alams, Dr. J. Wang and Dr. Keith Sharkey’s laboratory for technical assistance. This study was supported by grants from CIHR, CIHR Regenerative Medicine Initiative, and a fellowship to TC from the Sinneave Family Foundation.
Availability of data and materials
A single source file for all figure data has been included as Additional file 9.
TC: Designed and performed electrophysiological experiments, wrote the manuscript. JH: performed morphological experiments and some electrophysiological experiments. MDN: provided assistance in morphological experiments. BH: contributed to the design of experiments and edited the manuscript. All authors read and approved the final manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
- 29.Moyer J, Brown T. Patch-clamp techniques to brain slices. In: Walz W, Boulton AA, Baker GB, editors. Patch-clamp analysis: advanced techniques. Boston: Humana Press; 2002.Google Scholar
- 87.Zielinski BA, Prigge MB, Nielsen JA, Froehlich AL, Abildskov TJ, Anderson JS, et al. Longitudinal changes in cortical thickness in autism and typical development. Brain. 2013;137(Pt 6):1799–812.Google Scholar
- 90.Paxinos G, Tork I, Tecott LH. Atlas of the developing rat brain. San Diego: Academic Press; 1991.Google Scholar
- 95.Paxinos G, Watson C. The rat brain: in stereotaxic coordinates. 4th ed. New York: Academic Press; 1998.Google Scholar
- 106.Shepherd GM. Electrotonic properties of axons and dendrites. In: Zigmond MJ, Bloom FE, Landis SC, Roberts JL, Squire LR, editors. Fundamental neuroscience. San Diego: Academic Press; 1999. p. 107–28.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.