Dendritic spines and distributed circuits

Dendritic spines receive most excitatory connections in pyramidal cells and many other principal neurons. But why do neurons use spines, when they could accommodate excitatory contacts directly on their dendritic shafts? One suggestion is that spines serve to connect with passing axons, thus increasing the connectivity of the dendrites. Another hypothesis is that spines are biochemical compartments that enable input-specific synaptic plasticity. A third possibility is that spines have an electrical role, filtering synaptic potentials and electrically isolating inputs from each other. In this review, I argue that, when viewed from the perspective of the circuit function, these three functions dovetail with one another to achieve a single overarching goal: to implement a distributed circuit with widespread connectivity. Spines would endow these circuits with nonsaturating, linear integration and input-specific learning rules, which would enable them to function as neural networks, with emergent encoding and processing of information.     

Cortical area and species differences in dendritic spine morphology

Dendritic spines receive most excitatory inputs in the neocortex and are morphologically very diverse. Recent evidence has demonstrated linear relationships between the size and length of dendritic spines and important features of its synaptic junction and time constants for calcium compartmentalisation. Therefore, the morphologies of dendritic spines can be directly interpreted functionally. We sought to explore whether there were potential differences in spine morphologies between areas and species that could reflect potential functional differences. For this purpose, we reconstructed and measured thousands of dendritic spines from basal dendrites of layer III pyramidal neurons from mouse temporal and occipital cortex and from human temporal cortex. We find systematic differences in spine densities, spine head size and spine neck length among areas and species. Human spines are systematically larger and longer and exist at higher densities than those in mouse cortex. Also, mouse temporal spines are larger than mouse occipital spines. We do not encounter any correlations between the size of the spine head and its neck length. Our data suggests that the average synaptic input is modulated according to cortical area and differs among species. We discuss the implications of these findings for common algorithms of cortical processing.     

Class-specific features of neuronal wiring

Brain function relies on specificity of synaptic connectivity patterns among different classes of neurons. Yet, the substrates of specificity in complex neuropil remain largely unknown. We search for imprints of specificity in the layout of axonal and dendritic arbors from the rat neocortex. An analysis of 3D reconstructions of pairs consisting of pyramidal cells (PCs) and GABAergic interneurons (GIs) revealed that the layout of GI axons is specific. This specificity is manifested in a relatively high tortuosity, small branch length of these axons, and correlations of their trajectories with the positions of postsynaptic neuron dendrites. Axons of PCs show no such specificity, usually taking a relatively straight course through neuropil. However, wiring patterns among PCs hold a large potential for circuit remodeling and specificity through growth and retraction of dendritic spines. Our results define distinct class-specific rules in establishing synaptic connectivity, which could be crucial in formulating a canonical cortical circuit.     

Integration of subthreshold and suprathreshold excitatory barrages along the somatodendritic axis of pyramidal neurons

Neurons integrate inputs arriving in different cellular compartments to produce action potentials that are transmitted to other neurons. Because of the voltage- and time-dependent conductances in the dendrites and soma, summation of synaptic inputs is complex. To examine summation of membrane potentials and firing rates, we performed whole-cell recordings from layer 5 cortical pyramidal neurons in acute slices of the rat's somatosensory cortex. We delivered subthreshold and suprathreshold stimuli at the soma and several sites on the apical dendrite, and injected inputs that mimic synaptic barrages at individual or distributed sites. We found that summation of subthreshold potentials differed from that of firing rates. Subthreshold summation was linear when barrages were small but became supralinear as barrages increased. When neurons were discharging repetitively the rules were more diverse. At the soma and proximal apical dendrite summation of the evoked firing rates was predominantly sublinear whereas in the distal dendrite summation ranged from supralinear to sublinear. In addition, the integration of inputs delivered at a single location differed from that of distributed inputs only for suprathreshold responses. These results indicate that convergent inputs onto the apical dendrite and soma do not simply summate linearly, as suggested previously, and that distinct presynaptic afferents that target specific sites on the dendritic tree may perform unique sets of computations.     

Non-associative Potentiation of Perisomatic Inhibition Alters the Temporal Coding of Neocortical Layer 5 Pyramidal Neurons

In the neocortex, the coexistence of temporally locked excitation and inhibition governs complex network activity underlying cognitive functions, and is believed to be altered in several brain diseases. Here we show that this equilibrium can be unlocked by increased activity of layer 5 pyramidal neurons of the mouse neocortex. Somatic depolarization or short bursts of action potentials of layer 5 pyramidal neurons induced a selective long-term potentiation of GABAergic synapses (LTPi) without affecting glutamatergic inputs. Remarkably, LTPi was selective for perisomatic inhibition from parvalbumin basket cells, leaving dendritic inhibition intact. It relied on retrograde signaling of nitric oxide, which persistently altered presynaptic GABA release and diffused to inhibitory synapses impinging on adjacent pyramidal neurons. LTPi reduced the time window of synaptic summation and increased the temporal precision of spike generation. Thus, increases in single cortical pyramidal neuron activity can induce an interneuron-selective GABAergic plasticity effectively altering the computation of temporally coded information.     

Dense Neuron Clustering Explains Connectivity Statistics in Cortical Microcircuits

Local cortical circuits appear highly non-random, but the underlying connectivity rule remains elusive. Here, we analyze experimental data observed in layer 5 of rat neocortex and suggest a model for connectivity from which emerge essential observed non-random features of both wiring and weighting. These features include lognormal distributions of synaptic connection strength, anatomical clustering, and strong correlations between clustering and connection strength. Our model predicts that cortical microcircuits contain large groups of densely connected neurons which we call clusters. We show that such a cluster contains about one fifth of all excitatory neurons of a circuit which are very densely connected with stronger than average synapses. We demonstrate that such clustering plays an important role in the network dynamics, namely, it creates bistable neural spiking in small cortical circuits. Furthermore, introducing local clustering in large-scale networks leads to the emergence of various patterns of persistent local activity in an ongoing network activity. Thus, our results may bridge a gap between anatomical structure and persistent activity observed during working memory and other cognitive processes.     

Close homolog of L1 modulates area-specific neuronal positioning and dendrite orientation in the cerebral cortex

We show that the neural cell recognition molecule Close Homolog of L1 (CHL1) is required for neuronal positioning and dendritic growth of pyramidal neurons in the posterior region of the developing mouse neocortex. CHL1 was expressed in pyramidal neurons in a high-caudal to low-rostral gradient within the developing cortex. Deep layer pyramidal neurons of CHL1-minus mice were shifted to lower laminar positions in the visual and somatosensory cortex and developed misoriented, often inverted apical dendrites. Impaired migration of CHL1-minus cortical neurons was suggested by strikingly slower rates of radial migration in cortical slices, failure to potentiate integrin-dependent haptotactic cell migration in vitro, and accumulation of migratory cells in the intermediate and ventricular/subventricular zones in vivo. The restriction of CHL1 expression and effects of its deletion in posterior neocortical areas suggests that CHL1 may regulate area-specific neuronal connectivity and, by extension, function in the visual and somatosensory cortex.     

Dendritic spines of developing rat cortical neurons in culture

The formation of spines and their association with synapses were examined in developing cultured rat cortical neurons using fluorescence labeling techniques. Small protrusions were found on the processes of cultured cortical neurons after seven days in vitro (DIV), and the density of protrusions almost halved during the second week in vitro, after which it remained unchanged throughout the third week in vitro. The proportion of protrusions associated with the accumulation of the presynaptic marker, synaptophysin, increased steadily from <5% at 7 DIV to approximately 50% at 21 DIV. Based on the absence or presence of an enlargement at the end, protrusions on processes were further divided into filopodia and spines, respectively. The percentage of protrusions that were classified as spines increased steadily from approximately 5% at 3-4 DIV to approximately 80% at 18-20 DIV. The percentage of spines associated with synaptophysin accumulation increased gradually as the cortical neurons developed in vitro, reaching a plateau of approximately 40% after two weeks. However, the percentage of filopodia associated with synaptophysin accumulation never exceeded 5% during the first three weeks in vitro. Double-label staining the microfilaments and beta-tubulin or phosphorylated neurofilament H of cultured neurons further revealed many spines without any nearby axon-like processes. These findings suggest that spines are the dominant form of protrusion on the processes of more mature cortical neurons, that spines are the preferential sites where synapses reside, and that maintaining constant contact with axons is not essential for the formation of spines in cultured cortical neurons.     

Ultrastructure of the synapses of the initial axonal segment of the pyramidal neuron

Ultrastructure of the proximal part of the axon in the neurons, identified according to a number of morphological signs as pyramidal, has been studied in the layer III of the cat cerebral hemisphere sensomotor cortex. In sections, tangential to the cortical surface, in the initial axonal segment, a submembranous osmophilic layer and fasciculi of microtubules are revealed. On the initial segment spines are found, they contain cysterns resembling by their structure the spine system of the dendritic spines. Axonal terminals revealed along the axonal distribution are in contact both with the axonal trunk and with the spines. Regarding the initial segment, they are presynaptic, contain oval synaptic vesicles and form symmetric axo-axonal synapses only. In transversal sections axonal terminals are detected, arranging on the surface of the initial segment mostly as single ones, in longitudinal sections they are seen as clusters. Analysing the author's data and those from the literature, a conclusion is made that in intact animals the synaptic contacts at the initial segment of the axon are the only form of axo-axonal synapses in the neocortex.     

Destabilization of cortical dendrites and spines by BDNF.

Particle-mediated gene transfer and two-photon microscopy were used to monitor the behavior of dendrites of individual cortical pyramidal neurons coexpressing green fluorescent protein (GFP) and brain-derived neurotrophic factor (BDNF). While the dendrites and spines of neurons expressing GFP alone grew modestly over 24-48 hr, coexpressing BDNF elicited dramatic sprouting of basal dendrites, accompanied by a regression of dendritic spines. Compared to GFP-transfected controls, the newly formed dendrites and spines were highly unstable. Experiments utilizing Trk receptor bodies, K252a, and overexpression of nerve growth factor (NGF) demonstrated that these effects were mediated by secreted BDNF interacting with extracellular TrkB receptors. Thus, BDNF induces structural instability in dendrites and spines, which, when restricted to particular portions of a dendritic arbor, may help translate activity patterns into specific morphological changes.     

Area and layer patterning in the developing cerebral cortex

Two anatomical patterns characterize the neocortex, and both are essential for normal cortical function. First, neocortex is divided into anatomically distinct and functionally specialized areas that form a species-specific map. Second, neocortex is composed of layers that organize cortical connectivity. Recent studies of layer and area development have used time-lapse microscopy to follow cortical cell division and migration, gene arrays to find layer- or area- specific regulatory genes, time- and region- specific manipulations of candidate genes, and optical imaging to compare area maps in wild type with genetically altered mice. New observations clarify the molecular and cellular mechanisms that generate each pattern, and stress the links between layer and area formation.     

Inductive role of mossy fibers of hippocampus in the development of dendritic spines in aberrant synaptogenesis at neurotransplantation

The dentate fascia of the hippocampal formation isolated from 20-day-old Wistar rat fetuses was subjected to heterotopic transplantation into the somatosensory area of the neocortex of adult rats of the same strain. Five months after surgery, neurotransplantates, together with neighboring area of the neocortex, were studied using light and electron microscopy. We carried out a detailed study of the ultrastructure of the ectopic synaptic endings formed by the axons of granular neurons of the dentate fascia (mossy fibers) with neurons of the neocortex unusual for them in a normal state. Ultrastructural analysis revealed that most ectopic synaptic endings produce its determinant morphological features: giant sizes of presynaptic knobs, active zones with branched dendritic spines, and adherens junctions with the surface of dendrites. The data indicate that the mossy fibers growing from neurotransplantates induce structural and chemical reorganization of dendrites of the neocortex using transmembrane adherens junctions, such as puncta adherentia junctions. This results in the differentiation of active zones and development of dendritic spines typical for giant synaptic endings that are invaginated into presynaptic endings. Thus, the ability of neurons of the dentate fascia to form aberrant synaptic connections at transplantation results from the inductive synaptogenic properties of mossy fibers.     

Elimination of inhibitory synapses is a major component of adult ocular dominance plasticity

During development, cortical plasticity is associated with the rearrangement of excitatory connections. While these connections become more stable with age, plasticity can still be induced in the adult cortex. Here we provide evidence that structural plasticity of?inhibitory synapses onto pyramidal neurons is?a major component of plasticity in the adult neocortex. In?vivo two-photon imaging was used to monitor the formation and elimination of fluorescently labeled inhibitory structures on pyramidal neurons. We find that ocular dominance plasticity in the adult visual cortex is associated with rapid inhibitory synapse loss, especially of those present on dendritic spines. This occurs not only with monocular deprivation but also with subsequent restoration of binocular vision. We propose that in the adult visual cortex the experience-induced loss of inhibition may effectively strengthen specific visual inputs with limited need for rearranging the excitatory circuitry.     

How does the cerebral cortex work? Learning, attention, and grouping by the laminar circuits of visual cortex

The organization of neocortex into layers is one of its most salient anatomical features. These layers include circuits that form functional columns in cortical maps. A major unsolved problem concerns how bottom-up, top-down, and horizontal interactions are organized within cortical layers to generate adaptive behaviors. This article models how these interactions help visual cortex to realize: (i) the binding process whereby cortex groups distributed data into coherent object representations; (ii) the attentional process whereby cortex selectively processes important events; and (iii) the developmental and learning processes whereby cortex shapes its circuits to match environmental constraints. New computational ideas about feedback systems suggest how neocortex develops and learns in a stable way, and why top-down attention requires converging bottom-up inputs to fully activate cortical cells, whereas perceptual groupings do not.     

A model of the spatial-temporal characteristics of the alpha rhythm

A linear spatially distributed model of a chain of neurons and interneurons was investigated in relation to the generation of propagated alpha rhythmic activity. It was assumed that the elements of the chain were interconnected by means of recurrent collaterals and inhibitory fibres in such a way that the connectivity functions were assumed to be homogeneous and their strength was an exponentially decreasing function of distance. It was found that such a neuronal chain shows propagation properties for frequencies in the alpha band. The results obtained with the model are in agreement with the phase velocities encountered experimentally. In this way, it was possible to estimate the length of the neural fibres responsible for the phenomenon of propagated activity. The estimates obtained are in good agreement with recent quantitative neuroanatomical data on the circuitry of the neocortex.     

The microcephaly gene Donson is essential for progenitors of cortical glutamatergic and GABAergic neurons

Biallelic mutations in DONSON, an essential gene encoding for a replication fork protection factor, were linked to skeletal abnormalities and microcephaly. To better understand DONSON function in corticogenesis, we characterized Donson expression and consequences of conditional Donson deletion in the mouse telencephalon. Donson was widely expressed in the proliferation and differentiation zones of the embryonic dorsal and ventral telencephalon, which was followed by a postnatal expression decrease. Emx1-Cre-mediated Donson deletion in progenitors of cortical glutamatergic neurons caused extensive apoptosis in the early dorsomedial neuroepithelium, thus preventing formation of the neocortex and hippocampus. At the place of the missing lateral neocortex, these mutants exhibited a dorsal extension of an early-generated paleocortex. Targeting cortical neurons at the intermediate progenitor stage using Tbr2-Cre evoked no apparent malformations, whereas Nkx2.1-Cre-mediated Donson deletion in subpallial progenitors ablated 75% of Nkx2.1-derived cortical GABAergic neurons. Thus, the early telencephalic neuroepithelium depends critically on Donson function. Our findings help explain why the neocortex is most severely affected in individuals with DONSON mutations and suggest that DONSON-dependent microcephaly might be associated with so far unrecognized defects in cortical GABAergic neurons. Targeting Donson using an appropriate recombinase is proposed as a feasible strategy to ablate proliferating and nascent cells in experimental research.     

The development of peptide-containing neurons within neocortical transplants in adult mice

Transplantation of embryonic neocortex into adult host neocortex leads to the survival of many donor cells, with the subsequent differentiation of the cortical neurons within a loosely laminated cellular pattern. We wanted to know whether peptide-containing neurons that are known to exist in normal neocortex would survive in the transplants, and if so, whether they would differentiate into morphological cell types that normally contain these peptides in cortex. By 30 days after transplantation, the implants were well vascularized and the donor neurons appeared healthy in Nissl-stained preparations. AChE-positive axons grew across the interface and innervated the transplant in moderate densities. Immunocytochemical localization of peptides in the transplant revealed that processes containing the four peptides normally present in cortex also develop in the transplants. These were vasoactive intestinal polypeptide, cholecystokinin, pancreatic polypeptide and somatostatin. Other peptides not yet demonstrated in and presumably not present in neocortex, did not develop in the transplants. These included alpha-melanocyte stimulating hormone, arginine-vasopressin, corticotropin releasing factor, beta-endorphin and substance P. The results demonstrate that peptide-immunoreactive neurons survive in neural transplants, where they develop complicated patterns of axonal arborization. The conditions used in these experiments produced no evidence that peptidergic neurons within the transplant grow out of the transplant and into the host brain within six weeks. Similarly, host peptidergic axons were never seen crossing the interface zone and entering the transplant in any significant numbers.     

Visualization of Cortical Projection Neurons with Retrograde TET-Off Lentiviral Vector

We are interested in identifying and characterizing various projection neurons that constitute the neocortical circuit. For this purpose, we developed a novel lentiviral vector that carries the tetracycline transactivator (tTA) and the transgene under the TET Responsive Element promoter (TRE) on a single backbone. By pseudotyping such a vector with modified rabies G-protein, we were able to express palmitoylated-GFP (palGFP) or turboFP635 (RFP) in corticothalamic, corticocortical, and corticopontine neurons of mice. The high-level expression of the transgene achieved by the TET-Off system enabled us to observe characteristic elaboration of neuronal processes for each cell type. At higher magnification, we were able to observe fine structures such as boutons and spines as well. We also injected our retrograde TET-Off vector to the marmoset cortex and proved that it can be used to label the long-distance cortical connectivity of millimeter scale. In conclusion, our novel retrograde tracer provides an attractive option to investigate the morphologies of identified cortical projection neurons of various species.     

Maturation of a recurrent excitatory neocortical circuit by experience-dependent unsilencing of newly formed dendritic spines

Local recurrent excitatory circuits are ubiquitous in neocortex, yet little is known about their development or architecture. Here we introduce a quantitative technique for efficient single-cell resolution circuit mapping using 2-photon (2P) glutamate uncaging and analyze experience-dependent neonatal development of the layer 4 barrel cortex local excitatory circuit. We show that sensory experience specifically drives a 3-fold increase in connectivity at postnatal day (P) 9, producing a highly recurrent network. A profound dendritic spinogenesis occurs concurrent with the connectivity increase, but this is not experience dependent. However, in experience-deprived cortex, a much greater proportion of spines lack postsynaptic AMPA receptors (AMPARs) and synaptic connectivity via NMDA receptors (NMDARs) is the same as in normally developing cortex. Thus we describe a approach for quantitative circuit mapping and show that sensory experience sculpts an intrinsically developing template network, which is based on NMDAR-only synapses, by driving AMPARs into newly formed silent spines.