S1 microcircuits

• Dr. Dirk Feldmeyer, Research Centre Jülich, Institute of Neuroscience and Medicine, INM-2, Lo-Brandt-Str. D-52425 Jülich, Germany; RWTH Aachen University Hosital, Dept. of Psychiatry, Psychotherapy and Psychosomatics, Pauwelsstr. 30, D-52074, Aachen, Germany

The whisker-related portion of the primary somatosensory (S1) area (the ‘barrel cortex’) in rodents exhibits a topological arrangement that mirrors the peripheral (contralateral) tactile receptors, the vibrissae (whisker hairs) on the rodent’s snout. In cortical layer 4, barrel-like cytoarchitectonic units are discernible, each of which represents a single whisker hair (Woolsey and van der Loos, 1970). The extension of the borders of layer 4 (L4) barrels throughout all cortical layers has been taken to define a vertical barrel ‘column’ which is the anatomical correlate of a cortical column (Szentágothai, 1975). Because of this well-defined organisation, the barrel cortex has become a model system for investigating synaptic microcircuits and even long-range synaptic connectivity related to the structural representation of sensory receptors (for recent reviews see e.g. Fox, 2008; Bosman et al., 2011; Feldmeyer et al., 2013). This chapter will concentrate on the excitatory microcircuit in the S1 barrel cortex, for which a comprehensive structure and function relationship is now emerging.

Thalamocortical input to the barrel cortex

Sensory information detected by the whisker hairs on the rodent’s snout arrives in the barrel cortex via several different ‘pathways’. Each of these whisker-to-barrel cortex pathways consists of a four-neuron relay that links the sensory receptor with the S1 barrel cortex: a trigeminal ganglion cell, a trigeminothalamic neuron, a thalamocortical neuron and the intracortical target neurons (see the Scholarpedia chapter by Deschênes, 2009 for details). The most prominent parallel pathways are the so-called lemniscal, extralemniscal and paralemniscal pathways (Yu et al., 2006; Diamond et al., 2008; Bosman et al., 2011; see also Scholarpedia chapter by Deschênes, 2009). These pathways encode different aspects of vibrissae sensory signals. The lemniscal pathway runs through the principal trigeminal nucleus and projects to the dorsomedial (dm) portion of the ventroposterior medial (VPM) nucleus of the whisker-related thalamus; its cortical targets are described below. Cytoarchitectonic units analogous to the cortical L4 barrels can be found both in the principal trigeminal nucleus and the VPM where they are called barrelettes and barreloids, respectively. The lemniscal pathway combines both whisker motion (whisking) and object location (touch) information and conveys them to the S1 barrel cortex (Yu et al., 2006). The extralemniscal pathway relays whisker signals through the interpolar spinal trigeminal nucleus. Its’ trigeminothalamic afferents synapse in the ventrolateral portion of the VPM. Extralemniscal thalamocortical afferents have sparse projections to layers 3, 6 and the septa between the L4 barrels of S1 barrel cortex. However, they densely innervate layers 4 and 6 of the secondary somatosensory (S2) whisker-related cortex (see the Scholarpedia chapter by Deschênes, 2009 for details). The extralemniscal pathway conveys only touch signals to the S1 and S2 barrel-related cortex (Yu et al., 2006). Despite its functional relevance, this pathway will not be described in detail in this overview because little is known about their target S1 microcircuits. The paralemniscal pathway also runs through the interpolar spinal trigeminal nucleus and connects with the posterior medial thalamic nucleus of the thalamus (Pom); the intracortical targets of Pom afferents are described in detail below. In contrast to VPM, the Pom does not show cytoarchitectonic units such as barreloids. The paralemniscal pathway codes only whisker motion information (Yu et al., 2006). All layers of the barrel cortex receive excitatory synaptic input from at least one of these thalamic nuclei (Alloway, 2008; Meyer et al., 2010a; Oberlaender et al., 2012; Constantinople and Bruno, 2013). In this short review, the focus is on the intracortical sections of these pathways and their interactions.

Lemniscal thalamocortical targets

The major thalamorecipient layer in the somatosensory barrel cortex is layer 4 because it contains the highest density of thalamocortical (TC) axons per dendritic length (Jensen and Killackey, 1987; Chmielowska et al., 1989; Oberlaender et al., 2012). These TC afferents emanate from VPM and target both excitatory and inhibitory neurons in a L4 barrel (Bruno and Simons, 2002). In addition, VPM axons also innervate pyramidal cells and inhibitory GABAergic interneurons in layers 3, 5B and 6A (Meyer et al., 2010b; Meyer et al., 2011; Oberlaender et al., 2012). The majority of their boutons establish synapses onto excitatory neurons because the L4 excitatory neurons outnumber L4 interneurons by far (inhibitory/excitatory neuron ratio in layer 4 ~ 8% vs. 92%; Lefort et al., 2009; Meyer et al., 2011). Synaptic contacts formed by TC axons comprise only about 10-20% of the total number of synaptic contacts in layer 4 (White and Rock, 1979; Benshalom and White, 1986; Schoonover et al., 2014) and are therefore considerably outnumbered by intracortical synaptic connections. Under in vivo conditions, the average amplitude of unitary VPM-L4 spiny neuron EPSPs is ~1 mV suggesting that this synapse is of very low efficacy (Bruno and Sakmann, 2006). However, after whisker stimulation, VPM synaptic input to L4 spiny neurons is highly coincident and synchronous resulting in an efficient TC signal transfer (Jia et al., 2014; Schoonover et al., 2014) and does not require intralaminar amplification (Brumberg et al., 1999; Miller et al., 2001; Bruno and Sakmann, 2006).

Figure 1: ‘Canonical neuronal microcircuit’ as proposed by Douglas and Martin (1991).Three groups of neurons interact with each other: ‘smooth’ inhibitory cells, superficial excitatory neurons (P2+3) together with L4 excitatory neurons, and deep layer 5 and 6 neurons. Neurons from each group receive thalamic input, most prominently the superficial neurons. Note the strong recurrent connectivity in this microcircuit model. Dashed and thicker lines indicate weaker or stronger synaptic drive, respectively. Red, ‘smooth’ inhibitory cells; blue, excitatory connections and cells. Modified from Douglas and Martin (1991) with permission of John Wiley and Sons, Inc.

Paralemniscal thalamocortical targets

The major target regions of Pom (paralemniscal) TC input are layers 5A, 1 and 2. In addition, Pom axons also innervate the septa between layer 4 barrels (Alloway, 2008; Meyer et al., 2010b; Meyer et al., 2010a; Wimmer et al., 2010; Oberlaender et al., 2012) and layer 3 where they overlap with VPM axons. VPM and Pom inputs have been proposed to be the constituent elements of distinct intracortical columnar pathways, the ‘barrel column’ and the ‘septal column’ (Alloway, 2008). In layer 5A, Pom afferents probably establish synaptic contacts with basal dendrites of L5A pyramidal neurons (Petreanu et al., 2009) while in layer 1 and upper layer 2 they may target apical dendritic tufts of L2, L3 and L5 pyramidal neurons as well as L1 and L2 interneurons.

Excitatory S1 microcircuits

Excitation by the whisker-related thalamus is distributed within the barrel cortex via many distinct microcircuits. These microcircuits can be grouped into different subnetworks or microcircuits; however, these subnetworks are not separate and independent entities but interact at many different levels.

The first prototypical neuronal microcircuit that describes synaptic signalling from the sensory thalamus to and within a primary sensory cortex was the so-called ‘canonical microcircuit’ by Rodney Douglas and Kevan Martin for the primary visual (V1) cortex (Douglas and Martin, 1991; Binzegger et al., 2004; Douglas and Martin, 2004; Figure 1). It is worth noting that synaptic signalling shown here for the V1 cortex is highly recurrent and reciprocal within and between cortical layers. The morphological and synaptic properties of S1 neuronal microcircuits described in this review were to a large extent obtained from in vitro paired recording studies. An example of such a correlated structural-functional analysis is shown in Figure 2 for the intralaminar synaptic connection between two neighbouring L2/3 pyramidal cells (Feldmeyer et al., 2006; see also Table 1).

Figure 2: Characterization of a synaptic microcircuit in the S1 barrel cortex. (A) Electrophysiological recordings from synaptically coupled pyramidal cells in layer 2/3. A presynaptic action potential (red trace) elicits unitary EPSPs (grey traces) in the postsynaptic neuron. The bottom trace (black) shows the average time course of a unitary EPSP. (B) Distribution of the coefficient of EPSP amplitude variation (CV) for all recorded L2/3 pyramidal cell pairs. A low CV is indicative of a high reliability of a synaptic connection. (C) The EPSP vs. CV plot shows that even connections with small mean EPSP amplitudes can still reliable in S1 barrel cortex. (D) Biocytin filling of the synaptically coupled neuron pair shown in A. The connection had two synaptic contacts shown in E and F; E1 and F1 show electron microscopic verifications of these synaptic contacts; the small structures containing the white vesicles are the axon boutons. (G) Reconstruction of the L2/3 pyramidal cell pair shown in A and D. (G1) Enlarged dendritic domain of the postsynaptic pyramidal cell; the location of the two synaptic contacts is marked by magenta-coloured circles. Presynaptic dendrites, red; presynaptic axons, blue; postsynaptic dendrites, black; axons, green. Modified from Feldmeyer et al., 2006 with permission of John Wiley and Sons, Inc.

The ‘canonical’ S1 microcircuit

In this ‘canonical’ S1 microcircuit, afferents from the primary sensory thalamic nucleus target excitatory neurons in the granular and supragranular cortical layers. In addition, inhibitory interneurons receive thalamic input along with pyramidal cells in the infragranular layer, to a lesser extent. Thus, the signal flow in this microcircuit is from granular and supragranular layers to infragranular layers and subsequently to other cortical areas and subcortical brain regions. However, there is a substantial degree of intra- and translaminar feedback excitation.

Table 1: Characteristics of excitatory synaptic connections in a ‘canonical’ microcircuit’ of S1 barrel cortex.

Each synaptic connection is characterised by the apparent connectivity, the mean EPSP amplitude, the coefficient of EPSP amplitude variation (CV) and the failure rate (precentage of APs that did not elicit an EPSP). All synaptic parameters listed here were determined in the presence of 2 mM CaCl2 in the aCSF. L4 exc. neuron: L4 excitatory neuron, i.e. L4 spiny stellate, L4 star pyramid or L4 pyramidal cell. L2/3 PC, pyramidal cells in layer 2 and 3; tt L5 PC, thick-tufted L5 pyramidal cell.
* for this connection a release probability of 0.8 was determined (Silver et al., 2003).

Connection type	Apparent connectivity	mean EPSP amplitude	CV	Failure rate	Species	References
L4 exc. neuron → L4 exc. neuron	0.3	1.6±1.5 mV	0.37±0.16	5.3±7.8	Rat	Feldmeyer et al. 1999
	0.24	1.0±0.1 mV	n.d.	n.d.	Mouse	Lefort et al., 2009
L4 exc. neuron → L2/3 PC	~0.1-0.2	0.7±0.6 mV	0.27±0.13*	4.9±8.8%	Rat	Feldmeyer et al. 2002
	0.12 (L2)	1.0±0.2 mV (L2	n.d	n.d	Mouse	Lefort et al., 2009
	0.15 (L3)	0.6±0.1 mV (L3)	n.d	n.d
L2/3 PC → L2/3 PC	0.09 (local)	0.7±0.6 mV	n.d	n.d	Rat	Holmgren et al., 2003
	~0.1	1.0 ± 0.7 mV	0.33±0.18	3.2±7.8%	Rat	Feldmeyer et al. 2006
	0.09 (L2)	0.6±0.1 mV (L2)	n.d	n.d	Mouse	Lefort et al., 2009
	0.19 (L3)	0.8±0.1 mV (L3)	n.d	n.d
L2/3 PC → tt L5 PC	0.08 (L2)	0.2±0.0 mV (L2)	n.d	n.d	Mouse	Lefort et al., 2009
	0.12 (L3)	0.2±0.0 mV (L2)	n.d	n.d
tt L5 PC → tt L5 PC	~0.1	1.3±1.1 mV	0.52±0.41	14±7%	Rat	Markram et al., 1997
	-0.07	0.7±0.2 mV	n.d.	n.d.	Mouse	Lefort et al., 2009

In the S1 barrel cortex, a similar but not identical ‘canonical’ microcircuit exists which can - in approximation - be considered as intracortical part of the lemniscal pathway (Bureau et al., 2006). The main intracortical elements of this subnetwork are: L4 excitatory neurons (spiny stellates, star pyramids and pyramidal cells; Staiger et al., 2004; Oberlaender et al., 2012); L2 and L3 pyramidal cells and L5B pyramidal cells (Figure 3). Most studies on identified synaptic connections have concentrated on vertical, feed-forward signalling but a high degree of intralaminar reciprocal synaptic connections have also been identified. In addition to the vertical signal flow in a barrel column, there is mounting evidence for strong horizontal, transcolumnar synaptic signalling in the S1 barrel cortex and to other cortical areas (Zhang and Deschênes, 1997; Deschênes et al., 1998; Bruno et al., 2009; Narayanan et al., 2015). However, they will only be mentioned in passing as part of this review.

Figure 3: Synaptic connections in the ‘canonical’ microcircuit of S1 barrel cortex. (A) Simplified schematic drawing of synaptic connections in the synaptic connectivity in the S1 barrel cortex. Only L4 excitatory neurons (L4 ExcN: spiny stellate, star pyramids and pyramidal cells), L2, L3 and the thick-tufted (tt) L5B pyramidal cells are included here. (B) Morphological reconstructions of individual pairs of the excitatory synaptic connection depicted in (A): 1, L4-L4 connection, 2, L4-L2/3 connection, 3, L2/3-L2/3 connection, 4, L2-L5B connection and 5, L5B-L5B connection. Reconstructions were obtained from paired recordings and simultaneous biocytin fillings of synaptically coupled neurons in barrel cortex brain slices (B1-B5 modified from Feldmeyer et al., 1999; Silver et al., 2003; Feldmeyer et al., 2006; Reyes et al., 1999; Markram et al., 1997). L4 barrels are depicted in light grey; L2P, L3P, L2 and L3 pyramidal cells; L4ExcN, L4 excitatory neuron, ttL5BP, thick-tufted L5B pyramidal cell. Schematic barrels are depicted in light grey. Colour code as for Figure 2G.

In simplified terms, excitation arriving from the VPM results in strong recruitment of L4 excitatory neurons. L3 and L5B pyramidal cells are also recruited, but to a lesser degree. Within a L4 barrel, these neurons are recurrently interconnected with a high connectivity ratio of ~0.3 (see Table 1). Their translaminar targets are mainly pyramidal cells in supragranular layers (i.e. layers 2 and 3; but see below). In particular spiny stellate cells, the most numerous L4 cell type, show a largely ‘barrel column’-confined, vertical axonal projection and innervate pyramidal cells in both layer 2 and 3 (Lübke et al., 2000). L2 and L3 pyramidal cells innervate other pyramidal cells in their ‘home’ layer but provide also substantial synaptic input to thick-tufted pyramidal neurons in infragranular layer 5B. In turn, these L5B pyramidal cells provide output to subcortical target structures such as the thalamus, caudate-putamen, inferior colliculi and cerebellum (e.g. Zingg et al., 2014). The individual intracortical excitatory synaptic connections in this ‘canonical’ microcircuit in the S1 barrel cortex have been characterised in detail. These microcircuits comprise the synaptic connections between L4 excitatory neurons and L2/3 pyramidal cells, between L2/3 pyramidal cells and thick-tufted L5B pyramidal cells as well as the intralaminar reciprocal, recurrent connections between L4, L2/3 and L5B neurons (Markram et al., 1997; Feldmeyer et al., 1999; Feldmeyer et al., 2002; Holmgren et al., 2003; Silver et al., 2003; Feldmeyer et al., 2006; Lefort et al., 2009). Figure 3 shows a schematic wiring diagram (Figure 3A) and morphological reconstructions of these synaptic connections (Figure 3B). The majority shows a high release probability, as evidenced by a low EPSP failure rate in response to a presynaptic action potential (AP), a low variation in the EPSP amplitude (low coefficient of EPSP variation, c.v.) and a low paired pulse ratio (see Figure 2 for an example; Table 1). For the L4-L2/3 connections (Figure 3B2), this has been tested directly. In the presence of 2 mM extracellular Ca²⁺, the release probability of L4-L2/3 synapses was 0.8 (Silver et al., 2003). The number of synaptic contacts for individual connections varied between 2 and 8, the majority of which was located on the basal dendrites. This data shows that synaptic connections in the ‘canonical’ microcircuit are reliable, thereby ensuring an efficient sensory signal transfer in the barrel column. However, in vivo labelling of the axons of L2/3 and L5 pyramidal cell also revealed a strong and prominent horizontal projection domain both in layers 2, 3 and/or 5. These horizontal collaterals project across the entire barrel field (Bruno et al., 2009; Oberlaender et al., 2011; Narayanan et al., 2015), thereby integrating whisker-touch induced synaptic excitation in different barrel columns. In addition, there are long-range axon collaterals projecting to cortical areas outside the S1 barrel cortex that serve the interaction between is and other cortical areas. (e.g. the sensorimotor loop; see also Petreanu et al., 2007; Aronoff et al., 2010; Mao et al., 2011; Petreanu et al., 2012; Zingg et al., 2014).

Non-canonical, intracortical neuronal networks

Apart from the excitatory neurons in the ‘canonical’ microcircuit described above, there are several other synaptic connection types that do not fit this scheme. In general, these synaptic connections are part of the paralemniscal thalamic pathway from the Pom to the neocortex. However, the ‘paralemniscal’ intracortical microcircuit described here is not a segregated, independent pathway but highly interdigitated with the ‘canonical’, largely lemniscal pathway (Figure 4A). Constituent elements of the intracortical paralemniscal pathway are L5A pyramidal cells as well as the L2, L3 pyramidal cells and the apical tufts of L5B pyramidal cells, since these neurons are located in the major target regions of Pom afferents (Figure 4A; Bureau et al., 2006 see also Oberlaender et al., 2012). L4 septal neurons are also part of this microcircuit (see above; Wimmer et al., 2010). The major target neurons for the Pom afferents are the slender-tufted L5A pyramidal cells. These pyramidal cells have an extensive axonal projection in layers 1 and 2 (Figure 4B; Oberlaender et al., 2011). The Pom afferents will establish synaptic contacts not only with other L5A pyramidal cells (Figure 4C) and L2 pyramidal cells, but also the apical dendritic tufts of pyramidal cells in layer 3, 5A and 5B. This is of functional relevance because the dendritic tuft region of pyramidal cells has a high density of Ca²⁺ channels. This site has been shown to be the initiation zone for dendritic Ca²⁺ spikes (Figure 4A; Larkum et al., 1999; Larkum and Zhu, 2002; for reviews see Spruston, 2008; Larkum, 2013) whose activation is involved in coincidence detection mechanisms. Although the L5A-L2/3 pyramidal cell connectivity should be high (given their axodendritic domains; s. Figure 4B), Lefort and coworkers (Lefort et al., 2009) report only low connectivity ratios for supragranular projections (L3: 0.02, L2: 0.04). This is despite the strong synaptic input from these layers (L2: 0.1; L3: 0.06) and the fact that the L2 and L3 pyramidal cell axon density is comparatively lower in layer 5A. These conflicting results are most likely due to slice artifacts (i.e. the truncation of axonal collaterals) but also to the difficulty in detecting EPSPs from synaptic contacts on distal dendritic structures.

Figure 4: Synaptic connections in the intracortical paralemniscal microcircuit of S1 barrel cortex (A) Simplified schematic drawing of synaptic connections in the barrel cortex in the Pom pathway. Note that pyramidal cells in layers 2,3 and 5 receive thalamic input at their apical dendritic tuft where the Ca²⁺ spike initiation zone is located (marked by light red ellipse). L2 and L3 pyramidal cell are also likely to receive Pom input onto their basal dendrites. (B) Axonal projection pattern of L5A pyramidal cell filled in vivo revealing an extensive axonal collaterisation at the layer 1/layer 2 border. Modified from Oberlaender et al., 2011. (C) Synaptic connection between two L5A pyramidal cells. Modified from Frick et al., 2008. L2P, L3P, L5A, L5BP: pyramidal cells in layer 2, 3, 5A and 5B, respectively. Barrels are depicted in light grey. Colour code for reconstructions as in Figure 2G. Panel B reproduced with permission of the National Academy of Science of the USA.

The only synaptic connection in the ‘paralemniscal’ microcircuitry that has been studied in significant detail is the L5A-L5A connection (Figure 4C; Frick et al., 2008) which has relatively high connectivity ratio of ~0.2 (Lefort et al., 2009). It exhibits a low EPSP failure rate (2%) and CV (0.3) indicating that it similarly reliable as most other excitatory synaptic connections in the barrel cortex. Synaptic contacts are found on both basal dendrites and in the apical dendritic tuft. To date, the properties of translaminar connections in this ‘paralemniscal’ microcircuit are not known.

Early convergence of VPM and POm pathways in the S1 barrel cortex

As mentioned earlier, the lemniscal ‘canonical’ microcircuit and the paralemniscal microcircuit interact at several different stages. First of all, both thick-tufted L5B and L3 pyramidal cells receive monosynaptic and prominent TC input from both VPM (to their basal dendrites) and Pom (to their apical tufts, Figure 5A); this input is near-simultaneous. Further interaction occurs at the level of layer 4 and 5A which are the dominant target regions of VPM and Pom TC afferents, respectively. L4 excitatory neurons are also monosynaptically connected to L5A pyramidal cell with a connectivity ratio of 0.1. This connection shows a high reliability (similar to other excitatory connections in the S1 barrel cortex, see table 1); synaptic contacts are established on both basal and apical dendrites. This connection provides another short-latency link between the lemniscal and the paralemniscal pathways in the barrel cortex (Figure 5B; Feldmeyer et al., 2005; Bureau et al., 2006; Schubert et al., 2006; Lefort et al., 2009). The lemniscal and paralemniscal pathways also converge at several other stages of the neuronal network of the barrel column as shown in Figure 4A and Figure 5A, albeit disynaptically via slender-tufted L5A pyramidal cells and/or L4 excitatory neurons. For example, L5A pyramidal cells innervate apical tufts of L2, L3 and thick-tufted L5B pyramidal cells, all of which receive monosynaptic (as mentioned earlier) and/or disynaptic VPM input to their basal dendrites (via L4 excitatory neurons; Lefort et al., 2009; Feldmeyer et al., 2002; Staiger et al., 2014; Qi and Feldmeyer, unpublished) (Figure 5A). The near-coincident activation of basal dendrites and the apical dendritic tufts of these pyramidal cells through VPM and Pom synaptic input has been suggested to play a role during sensorimotor behavioural paradigms, such as object location during active whisking (Ahissar et al., 2000; Oberlaender et al., 2011). Alternatively, this ‘sensorimotor’ coincidence detection may result from simultaneous synaptic input from both the primary motor cortex and the VPM (Xu et al., 2012).

Figure 5: Interdigitation of lemniscal (VPM) and paralemniscal (Pom) intracortical pathways (A) Simplified schematic drawing of synaptic connections showing interaction between the two TC pathways. L4 excitatory neurons which receive VPM TC input are connected to excitatory neurons in all cortical layers although the connectivity in granular and supragranular layers is high than in infragranular layers. They interdigitate with L5A pyramidal cells which are the major targets of the Pom TC input. In addition, L5A pyramidal cells innervate also L5B pyramidal cells, which are also target neurons of the VPM TC input on both basal and apical dendrites. Finally, L4 excitatory neurons innervate L2 and L3 pyramidal cells, mostly on their basal dendrites; both neuron types receive either direct or indirect Pom input via Pom afferents and/or L5A pyramidal cell axons. (B) Synaptically coupled pair of a L4 spiny stellate neurons and a slender-tufted L5A pyramidal cell. This synaptic connection represents an early, short latency link between the VPM and Pom TC input into the neocortex. L4 barrels are depicted in light grey; L2P, L3P, L2 and L3 pyramidal cells; L4ExcN, L4 excitatory neuron, ttL5BP, thick-tufted L5B pyramidal cell. Colour code for reconstruction as in Figure 2G.

Corticothalamic projections

The S1 barrel cortex contains also several synaptic microcircuits that provide TC-CT feedback. The central elements of this feedback circuit are the L6 corticothalamic (CT) pyramidal cells that receive strong and depressing synaptic input from VPM ( Figure 6A; Beierlein and Connors, 2002; Cruikshank et al., 2010). L6A CT axon project back mainly to the same thalamic nucleus; however, projections to Pom as well as to both of these thalamic nuclei have been described (Figure 6A; s. Zhang and Deschênes, 1997; Deschênes et al., 1998). In contrast to pyramidal cells from other cortical layers, the majority of L6A pyramidal cells have short apical dendrites with sparse or even no apical tufts terminating predominantly in layers 3 to 5A. L6A pyramidal cells that provide only CT to VPM are predominantly located in the upper half of layer 6. Most L6A CT pyramidal cells have rather short axons - even when in vivo dye fillings are used - with a narrow axonal domain that is almost confined to a ‘barrel column’. In marked contrast to L6A CT pyramidal cells, corticocortically (CC) projecting pyramidal cells in layer 6A have extensive, profusely branching axons and send collaterals to other cortical areas such as the S2 cortex or motor cortex but have no obvious subcortical target (Zhang and Deschênes, 1997; Kumar and Ohana, 2008; Chen et al., 2009; Tanaka et al., 2011; Pichon et al., 2012). In general, L6 pyramidal cells have been reported to receive synaptic input only from layers 4, 5 and 6; the synaptic connectivity is reportedly low for these pathways, ranging from 0.03 to 0.06 (Lefort et al., 2009). Thus, apart from direct, monosynaptic TC VPM input, L4 excitatory neurons and thick-tufted L5B pyramidal cells provide disynaptic VPM input and thus serve as intracortical elements of a TC-CT feedback loop (Figure 6B). Translaminar synaptic input showed paired pulse EPSP depression indicative of a high release probability (Qi and Feldmeyer, 2015; see also Mercer et al., 2005; Thomson, 2010). L6A CC and CT pyramidal project also back to layers 5 and 6 (Mercer et al., 2005; Thomson, 2010). In addition, L6 projections to L4 excitatory neurons have been identified in the barrel cortex (Qi et al., 2015, cf. Stratford et al., 1996 for V1 cortex). These connections are unique because they show EPSP facilitation (Qi et al., 2015) in contrast to the paired-pulse depression observed for most other TC and CC connections in the S1 barrel cortex. Thus, L4 excitatory neurons and L6 pyramidal cells also provide a secondary, intracortical feedback control with the L4-L6 connection showing a delayed recruitment because of the paired pulse depression.

Figure 6: Synaptic microcircuitry in corticothalamic connections. Simplified schematic drawing of synaptic connections that play a role in corticothalamic feedback. VPM and its primary intracortical target neurons are shown in red, Pom and its primary cortical target in magenta. Corticothalamic output pathways is coloured in violet. L6A CT pyramidal cells provide synaptic input to the same thalamic nucleus from which they receive synaptic input; in contrast, L5B pyramidal cells receive mainly VPM TC input but project to Pom. L4 excitatory neurons interact with L6A pyramidal cells in an intracortical feedback loop (see text for details). (B) Synaptically coupled pair of a presynaptic L4 spiny stellate neuron and a L6A corticothalamic pyramidal cell. L4 barrels are depicted in light grey; L4ExcN, L4 excitatory neuron, L5AP, L5A pyramidal cell; L5BP, thick-tufted L5B pyramidal cell; CT L6AP, L6A CT pyramidal cell. Colour code for reconstruction as in Figure 2G.

It has been suggested, that the L4 input to L6 CT pyramidal cells is strong and focussed to its home ‘barrel column’, indicating that neurons in this layer are involved in shaping the cortical modulation of activity in the somatosensory thalamus (Tanaka et al., 2011; Qi and Feldmeyer, 2015). Notably, most synaptic contacts established by L4 excitatory neurons are on apical tufts of L6A pyramidal cells (Qi and Feldmeyer, 2015). Under this condition dendritic filtering is substantial, which may explain the low L4-L6 connectivity ratio found in in vitro experiments (Lefort et al., 2009). In addition to the L4 input, a strong synaptic input from L6 itself is likely to exist because CC projecting L6 pyramidal cells have extensive, profusely branching axons that project to other cortical areas such as the S2 cortex or the motor cortex, without obvious subcortical target structures. The connection probability is comparable to other intracortical excitatory connections (Mercer et al., 2005; Qi and Feldmeyer, unpublished; but see Lefort et al., 2009 for lower connectivity ratios). This is another example of a distinct, apical dendritic tuft targeting and basal dendrite targeting synaptic arrangement (as found in VPM and Pom innervation or L4 and L5A input to basal and apical dendritic tufts, respectively, of L5B pyramidal cells) and may serve in the detection of coincident synaptic stimuli. Another neuronal population involved in CT signalling is a subpopulation of thick-tufted L5B pyramidal cells that is innervated by VPM afferents but projects back to Pom. In this case the CT projection does not target the same thalamic nucleus from which it receives TC input, as is the case for L6A CT pyramidal cells. These L5B CT pyramidal cells may serve as a feedback control of the VPM over the Pom in a VPM-L5B-Pom loop. Thick-tufted L5B pyramidal cell axons receive VPM input (see above) and generate one or two clusters of large diameter (2-8 µm) presynaptic boutons in Pom (Hoogland et al., 1991; Bourassa et al., 1995; Groh et al., 2008; Groh et al., 2013). L5B-Pom synapses show a high release probability of 0.8 and are therefore very efficacious. Single unitary EPSPs can elicit several APs in the thalamic relay neurons thereby acting as ‘drivers’ of the Pom (Groh et al., 2008). However, spontaneous activity of the L5B pyramidal cells significantly reduces this ‘driving’ action through a strong short-term synaptic depression. Because of this it has been suggested that the L5B-Pom synapse works in two functional modes. When the spontaneous activity of this synapse is high, only synchronous activity of several L5B inputs will induce spiking of the thalamic neurons. Therefore, the synapse will act as a coincidence detector. When the spontaneous activity is sparse (which is the case during active whisking or cortical silence) a single L5B input will ‘fire’ the postsynaptic Pom neuron. Thus, the degree of spontaneous activity determines whether the CT L5B-Pom synapse acts as a detector of synchronous neuronal activity or of cortical silence (Groh et al., 2008). According to another hypothesis the L5B-Pom connection is part of a feed-forward, trans-thalamic signalling pathway from VPM via L5B pyramidal cells of barrel cortex to Pom and from where it ‘drives’ higher order cortical areas, e.g. the S2 cortex (see e.g. Killackey and Sherman, 2003; Theyel et al., 2010; Sherman and Guillery, 2011; for a review see Guillery and Sherman, 2011). However, it is likely that the L5B CT pyramidal cells are elements in both the feed-forward and the feedback pathways described above.

Conclusion

It should be noted that this review provides a simplified concept of the neuronal microcircuits in the S1 barrel cortex. The excitatory S1 microcircuits described here are certainly not separate subnetworks but synaptically interact at several distinct stages; thus, further complicating the understanding of their functional properties. It is likely that with our increasing knowledge of the morphological, electrophysiological and synaptic data from both in vivo and in vitro experiments, we will understand more about the structure-function relationship of barrel cortex synaptic microcircuit, both within a ‘barrel column’ and beyond.

References

• Alloway, K D (2008). Information processing streams in rodent barrel cortex: The differential functions of barrel and septal circuits. Cerebral Cortex 18: 979-989.
• Aronoff, R et al. (2010). Long-range connectivity of mouse primary somatosensory barrel cortex. European Journal of Neuroscience 31: 2221-2233.
• Beierlein, M and Connors, B W (2002). Short-term dynamics of thalamocortical and intracortical synapses onto layer 6 neurons in neocortex. Journal of Neurophysiology 88: 1924-1932.
• Benshalom, G and White, E L (1986). Quantification of thalamocortical synapses with spiny stellate neurons in layer IV of mouse somatosensory cortex. Journal of Comparative Neurology 253: 303-314.
• Bosman, L W et al. (2011). Anatomical pathways involved in generating and sensing rhythmic whisker movements. Frontiers in Integrative Neuroscience 5: 53.
• Bourassa, J; Pinault, D and Deschênes, M (1995). Corticothalamic projections from the cortical barrel field to the somatosensory thalamus in rats: a single-fibre study using biocytin as an anterograde tracer. European Journal of Neuroscience 7: 19-30.
• Brumberg, J C; Pinto, D J and Simons, D J (1999). Cortical columnar processing in the rat whisker-to-barrel system. Journal of Neurophysiology 82: 1808-1817.
• Bruno, R M and Sakmann, B (2006). Cortex is driven by weak but synchronously active thalamocortical synapses. Science 312: 1622-1627.
• Bruno, R M; Hahn, T T; Wallace, D J; de Kock, C P and Sakmann, B (2009). Sensory experience alters specific branches of individual corticocortical axons during development. The Journal of Neuroscience 29: 3172-3181.
• Bureau, I; von Saint Paul, F and Svoboda, K (2006). Interdigitated paralemniscal and lemniscal pathways in the mouse barrel cortex. PLoS Biology 4: e382.
• Chen, C C; Abrams, S; Pinhas, A and Brumberg, J C (2009). Morphological heterogeneity of layer VI neurons in mouse barrel cortex. Journal of Comparative Neurology 512: 726-746.
• Chmielowska, J; Carvell, G E and Simons, D J (1989). Spatial organization of thalamocortical and corticothalamic projection systems in the rat SmI barrel cortex. Journal of Comparative Neurology 285: 325-338.
• Constantinople, C M and Bruno, R M (2013). Deep cortical layers are activated directly by thalamus. Science 340: 1591-1594.
• Cruikshank, S J; Urabe, H; Nurmikko, A V and Connors, B W (2010). Pathway-specific feedforward circuits between thalamus and neocortex revealed by selective optical stimulation of axons. Neuron 65: 230-245.
• Deschênes, M (2009). Vibrissal afferents from trigeminus to cortices. Scholarpedia 4(5): 7454. http://www.scholarpedia.org/article/Vibrissal_afferents_from_trigeminus_to_cortices. (see also pages mmm-nnn of this book)
• Deschênes, M; Veinante, P and Zhang, Z W (1998). The organization of corticothalamic projections: Reciprocity versus parity. Brain Research. Brain Research Reviews 28: 286-308.
• Douglas, R J and Martin, K A (1991). A functional microcircuit for cat visual cortex. Journal of Physiology 440: 735-769.
• Douglas, R J and Martin, K A (2004). Neuronal circuits of the neocortex. Annual Review of Neuroscience 27: 419-451.
• Feldmeyer, D; Roth, A and Sakmann, B (2005). Monosynaptic connections between pairs of spiny stellate cells in layer 4 and pyramidal cells in layer 5A indicate that lemniscal and paralemniscal afferent pathways converge in the infragranular somatosensory cortex. The Journal of Neuroscience 25: 3423-3431.
• Feldmeyer, D; Lübke, J and Sakmann, B (2006). Efficacy and connectivity of intracolumnar pairs of layer 2/3 pyramidal cells in the barrel cortex of juvenile rats. Journal of Physiology 575: 583-602.
• Feldmeyer, D; Egger, V; Lübke, J and Sakmann, B (1999). Reliable synaptic connections between pairs of excitatory layer 4 neurones within a single 'barrel' of developing rat somatosensory cortex. Journal of Physiology 521(Pt 1): 169-190.
• Feldmeyer, D; Lübke, J; Silver, R A and Sakmann, B (2002). Synaptic connections between layer 4 spiny neurone-layer 2/3 pyramidal cell pairs in juvenile rat barrel cortex: Physiology and anatomy of interlaminar signalling within a cortical column. The Journal of Physiology 538: 803-822.
• Feldmeyer, D et al. (2013). Barrel cortex function. Progress in Neurobiology 103: 3-27.
• Fox, K D (2008). Barrel Cortex, 1st Edition. Cambridge, UK: Cambridge University Press.
• Frick, A; Feldmeyer, D; Helmstaedter, M and Sakmann, B (2008). Monosynaptic connections between pairs of L5A pyramidal neurons in columns of juvenile rat somatosensory cortex. Cerebral Cortex 18: 397-406.
• Groh, A; de Kock, C P; Wimmer, V C; Sakmann, B and Kuner, T (2008). Driver or coincidence detector: Modal switch of a corticothalamic giant synapse controlled by spontaneous activity and short-term depression. The Journal of Neuroscience 28: 9652-9663.
• Groh, A et al. (2013). Convergence of cortical and sensory driver inputs on single thalamocortical cells. Cerebral Cortex.
• Guillery, R W and Sherman, S M (2011). Branched thalamic afferents: What are the messages that they relay to the cortex? Brain Research Reviews 66: 205-219.
• Holmgren, C; Harkany, T; Svennenfors, B and Zilberter, Y (2003). Pyramidal cell communication within local networks in layer 2/3 of rat neocortex. Journal of Physiology 551: 139-153.
• Hoogland, P V; Wouterlood, F G; Welker, E and van der Loos, H (1991). Ultrastructure of giant and small thalamic terminals of cortical origin: a study of the projections from the barrel cortex in mice using Phaseolus vulgaris leucoagglutinin (PHA-L). Experimental Brain Research 87: 159-172.
• Jensen, K F and Killackey, H P (1987). Terminal arbors of axons projecting to the somatosensory cortex of the adult rat. I. The normal morphology of specific thalamocortical afferents. The Journal of Neuroscience 7: 3529-3543.
• Jia, H; Varga, Z; Sakmann, B and Konnerth, A (2014). Linear integration of spine Ca²⁺ signals in layer 4 cortical neurons in vivo. Proceedings of the National Academy of Sciences of the United States of America 111: 9277-9282.
• Killackey, H P and Sherman, S M (2003). Corticothalamic projections from the rat primary somatosensory cortex. The Journal of Neuroscience 23: 7381-7384.
• Kumar, P and Ohana, O (2008). Inter- and intralaminar subcircuits of excitatory and inhibitory neurons in layer 6a of the rat barrel cortex. Journal of Neurophysiology 100: 1909-1922.
• Larkum, M (2013). A cellular mechanism for cortical associations: An organizing principle for the cerebral cortex. Trends in Neuroscience 36: 141-151.
• Larkum, M E and Zhu, J J (2002). Signaling of layer 1 and whisker-evoked Ca²⁺ and Na⁺ action potentials in distal and terminal dendrites of rat neocortical pyramidal neurons in vitro and in vivo. The Journal of Neuroscience 22: 6991-7005.
• Larkum, M E; Zhu, J J and Sakmann, B (1999). A new cellular mechanism for coupling inputs arriving at different cortical layers. Nature 398: 338-341.
• Lefort, S; Tomm, C; Floyd Sarria, J C and Petersen, C C (2009). The excitatory neuronal network of the C2 barrel column in mouse primary somatosensory cortex. Neuron 61: 301-316.
• Lübke, J; Egger, V; Sakmann, B and Feldmeyer, D (2000). Columnar organization of dendrites and axons of single and synaptically coupled excitatory spiny neurons in layer 4 of the rat barrel cortex. The Journal of Neuroscience 20: 5300-5311.
• Mao, T et al. (2011). Long-range neuronal circuits underlying the interaction between sensory and motor cortex. Neuron 72: 111-123.
• Markram, H; Lübke, J; Frotscher, M; Roth, A and Sakmann, B (1997). Physiology and anatomy of synaptic connections between thick tufted pyramidal neurones in the developing rat neocortex. Journal of Physiology 500(Pt 2): 409-440.
• Mercer, A et al. (2005). Excitatory connections made by presynaptic cortico-cortical pyramidal cells in layer 6 of the neocortex. Cerebral Cortex 15: 1485-1496.
• Meyer, H S et al. (2010a). Number and laminar distribution of neurons in a thalamocortical projection column of rat vibrissal cortex. Cerebral Cortex 20: 2277-2286.
• Meyer, H S et al. (2011). Inhibitory interneurons in a cortical column form hot zones of inhibition in layers 2 and 5A. Proceedings of the National Academy of Sciences of the United States of America 108: 16807-16812.
• Meyer, H S et al. (2010b). Cell type-specific thalamic innervation in a column of rat vibrissal cortex. Cerebral Cortex 20: 2287-2303.
• Miller, K D; Pinto, D J and Simons, D J (2001). Processing in layer 4 of the neocortical circuit: New insights from visual and somatosensory cortex. Current Opinion in Neurobiology 11: 488-497.
• Oberlaender, M et al. (2011). Three-dimensional axon morphologies of individual layer 5 neurons indicate cell type-specific intracortical pathways for whisker motion and touch. Proceedings of the National Academy of Sciences of the United States of America 108: 4188-4193.
• Oberlaender, M et al. (2012). Cell type-specific three-dimensional structure of thalamocortical circuits in a column of rat vibrissal cortex. Cerebral Cortex 22: 2375-2391.
• Petreanu, L; Huber, D; Sobczyk, A and Svoboda, K (2007). Channelrhodopsin-2-assisted circuit mapping of long-range callosal projections. Nature Neuroscience 10: 663-668.
• Petreanu, L et al. (2012). Activity in motor-sensory projections reveals distributed coding in somatosensation. Nature 489: 299-303.
• Pichon, F; Nikonenko, I; Kraftsik, R and Welker, E (2012). Intracortical connectivity of layer VI pyramidal neurons in the somatosensory cortex of normal and barrelless mice. European Journal of Neuroscience 35: 855-869.
• Qi, G and Feldmeyer, D (2015). Dendritic target region-specific formation of synapses between excitatory layer 4 neurons and layer 6 pyramidal cells. Cerebral Cortex Epub.
• Qi, G; Radnikow, G and Feldmeyer, D (2015). Electrophysiological and morphological characterization of neuronal microcircuits in acute brain slices using paired patch-clamp recordings. Journal of Visualized Experiments e52358.
• Schoonover, C E et al. (2014). Comparative strength and dendritic organization of thalamocortical and corticocortical synapses onto excitatory layer 4 neurons. The Journal of Neuroscience 34: 6746-6758.
• Schubert, D; Kötter, R; Luhmann, H J and Staiger, J F (2006). Morphology, electrophysiology and functional input connectivity of pyramidal neurons characterizes a genuine layer Va in the primary somatosensory cortex. Cerebral Cortex 16: 223-236.
• Sherman, S M and Guillery, R W (2011). Distinct functions for direct and transthalamic corticocortical connections. Journal of Neurophysiology 106: 1068-1077.
• Silver, R A; Lübke, J; Sakmann, B and Feldmeyer, D (2003). High-probability uniquantal transmission at excitatory synapses in barrel cortex. Science 302: 1981-1984.
• Spruston, N (2008). Pyramidal neurons: Dendritic structure and synaptic integration. Nature Reviews Neuroscience 9: 206-221.
• Staiger, J F; Bojak, I; Miceli, S and Schubert, D (2015). A gradual depth-dependent change in connectivity features of supragranular pyramidal cells in rat barrel cortex. Brain Structure & Function 220(3): 1317-1337.
• Staiger, J F et al. (2004). Functional diversity of layer IV spiny neurons in rat somatosensory cortex: quantitative morphology of electrophysiologically characterized and biocytin labeled cells. Cerebral Cortex 14: 690-701.
• Stratford, K J; Tarczy-Hornoch, K; Martin, K A; Bannister, N J and Jack, J J (1996). Excitatory synaptic inputs to spiny stellate cells in cat visual cortex. Nature 382: 258-261.
• Szentágothai, J (1975). The 'module-concept' in cerebral cortex architecture. Brain Research 95: 475-496.
• Tanaka, Y R et al. (2011). Local connections of excitatory neurons to corticothalamic neurons in the rat barrel cortex. The Journal of Neuroscience 31: 18223-18236.
• Theyel, B B; Llano, D A and Sherman, S M (2010). The corticothalamocortical circuit drives higher-order cortex in the mouse. Nature Neuroscience 13: 84-88.
• Thomson, A M (2010). Neocortical layer 6, a review. Frontiers in Neuroanatomy 4: 13.
• White, E L and Rock, M P (1979). Distribution of thalamic input to different dendrites of a spiny stellate cell in mouse sensorimotor cortex. Neuroscience Letters 15: 115-119.
• Wimmer, V C; Bruno, R M; de Kock, C P; Kuner, T and Sakmann, B (2010). Dimensions of a projection column and architecture of VPM and POm axons in rat vibrissal cortex. Cerebral Cortex 20: 2265-2276.
• Zhang, Z W and Deschênes, M (1997). Intracortical axonal projections of lamina VI cells of the primary somatosensory cortex in the rat: A single-cell labeling study. The Journal of Neuroscience 17: 6365-6379.
• Zingg, B et al. (2014). Neural networks of the mouse neocortex. Cell 156: 1096-1111.

Sponsored by: Prof. Tony J. Prescott, Dept Psychology, Univ of Sheffield, UK

Reviewed by: Dr. Shruti Muralidhar, University of Utah, Salt Lake City, Utah, U.S.A

Reviewed by: Prof. Ehud Ahissar, Department of Neurobiology, The Weizmann Institute, Rehovot

Reviewed by: Dr. Christiaan de Kock, Dept. Of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, Amsterdam, The Netherlands

Accepted on: 2015-05-09 22:01:43 GMT