Elsevier

Neural Networks

Volume 17, Issues 5–6, June–July 2004, Pages 633-646
Neural Networks

2004 Special Issue
Electrophysiological classes of neocortical neurons

https://doi.org/10.1016/j.neunet.2004.04.003Get rights and content

Abstract

Neocortical network behavior and neocortical function emerge from synaptic interactions among neurons with specific electrophysiological and morphological characteristics. The intrinsic electrophysiological properties of neurons define their firing patterns and their input–output functions with critical consequences for their functional properties within the network. Understanding the role played by the active non-linear properties caused by ionic conductances distributed in the soma and the dendrites is a critical step towards understanding cortical function. Here I present a brief description of electrophysiological and morphological characteristics of neocortical cells that allow their classification in categories. I review some examples of differences in functional properties among different electrophysiological cell classes in the visual cortex, as well as the role played by specific ionic conductances in defining firing and accommodation properties of neocortical neurons.

Introduction

This review examines the proposal that understanding the electrophysiological properties of neocortical neurons is a key element in understanding neocortical function. Such a proposition is based on the following rationale: (i) most of the informational output of an individual neuron is encoded in some combination of the rate and the timing of its action potentials, and (ii) the spike output of neurons is strongly shaped by their intrinsic electrophysiological properties.

Two main factors determine neuronal electrophysiological properties: (i) the types of voltage sensitive channels expressed by the neuron and (ii) their density and distribution over the soma-dendritic membrane. Voltage dependent conductances impose non-linearities in the current–voltage relation of neurons that strongly modify their responsiveness with critical functional consequences. In addition, intrinsic electrophysiological properties shape the spike output by dynamically interacting with postsynaptic potentials or strongly determining it as in the case of pacemaker cells (Llinas, 1988). Finally, electrophysiological properties and the resulting firing patterns are uniform enough (but see Steriade, 2004a) to allow the definition of electrophysiological classes (Connors and Gutnick, 1990, Llinas, 1988, Nowak et al., 2003, Nunez et al., 1993, Steriade, 2004a).

Early models of cortical function relied entirely on complex connectivity patterns among cells with similar electroresponsive properties. It has become clear, however, that such models do not capture the dynamic properties of cortical neurons and cortical networks (Amitai and Connors, 1995b, Connors and Gutnick, 1990, Hausser et al., 2000, Llinas, 1988, Marder, 1998, Traub et al., 2003, Traub et al., 1996). The variety of cellular morphologies, neurotransmitters, connectivity patterns, specific synaptic dynamics and electrophysiological properties of cortical cells seems to indicate that an adequate representation of cortical operations should include more than average spike rates and connectivity patterns. Here I present a brief account of the electrophysiological cell classes in neocortex, their morphological characteristics, and discuss a few examples of how their intrinsic properties may affect input–output transformations.

In the early years of the study of electroresponsiveness of neurons in the mammalian central nervous system, following the first intracellular recordings in vivo from spinal motorneurons in the anesthetized cat (Brock, Coombs, & Eccles, 1952), a series of studies showed that spikes are generated in the axon hillock-initial segment by the constant fluctuations of the membrane potential around a fixed threshold (Eccles, 1957; spike initiation in the axon hillock-initial segment has recently been demonstrated using patch clamp technique in slices, Colbert and Johnston, 1996, Stuart and Sakmann, 1994). Those early studies proposed that the variations in membrane potential were determined by the linear processes of synaptic integration, which consisted of the spatial and temporal summation of inputs along the somatodendritic membrane (Eccles, 1957, Eccles, 1964). This view received a rigorous mathematical formulation in the work of Rall (1957) and was extended to all neurons of the central nervous system, including of course the cerebral cortex (Calvin, 1975). According to this early view, all central nervous system neurons had similar electroresponsive properties and thus, complexity and function emerged from neuronal interactions within large networks.

That not all neurons shared the same electrophysiological characteristics was already clear in the extracellular recordings from neocortex of Mountcastle and colleagues, who distinguished between cells with ‘thin’ spikes and high spontaneous firing rates and cells with longer spikes and showed lower sustained firing rates that were termed ‘regular spiking’ (Mountcastle, Talbot, Sakata, & Hyvarinen, 1969). A distinction was also made among pyramidal cells projecting in the pyramidal tract. Those with fast conduction velocities have very short spikes (<1 ms) and low accommodation rates; those with slow conduction velocities have longer spikes (∼1 ms) and high accommodation rates (Calvin and Sypert, 1976, Deschenes et al., 1979).

Another important challenge to the early ideas came from studies of dendritic excitability. As early as 1958, intracellular recordings from chromatolyzed motorneurons (cells in which the axon had been cut) suggested that dendrites could generate spikes (Eccles, Libet, & Young, 1958). This notion was supported by intracellular recordings of fast prepotentials, presumably originating in the dendrites, from the somas of CA1 hippocampal pyramidal cells (Kandel & Spencer, 1961). Definitive proof that dendrites could generate autoregenerative potentials, which are based on Na+ and Ca2+ processes, was given by direct intracellular recordings of dendrites in Purkinje cells (Llinas & Nicholson, 1971), cortical pyramidal cells (Houchin, 1973) and hippocampal cells (Benardo et al., 1982, Wong et al., 1979). In neocortical pyramidal cells, a series of studies using different technical approaches such as intradendritic recordings in vivo (Pockberger, 1991), and in vitro (Amitai, Friedman, Connors, & Gutnick, 1993), whole cell patch recordings in vitro from dendrites (Kim and Connors, 1993, Larkum et al., 1999, Stuart and Sakmann, 1994) and from soma (Schwindt and Crill, 1999, Schwindt and Crill, 1997), and fluorescence imaging using Ca2+ indicators (Markram and Sakmann, 1994, Oakley et al., 2001, Yuste et al., 1994) demonstrated that dendrites are capable of generating both Na+ and Ca2+ action potentials. Thus, a new view emerged in which active dendritic excitability, superimposed on basic electrotonic properties, would modify postsynaptic potentials to produce a dynamic and regulated integration of synaptic input (for a review see Johnston et al., 1996, Llinas, 1975, Reyes, 2001).

A heated debate in the matter of neuronal responsiveness was generated when the first intracellular recordings in the thalamus (Andersen and Eccles, 1962, Purpura and Cohen, 1962) revealed an electrophysiological behavior that could not be explained by the current models of neuronal excitability. A depolarizing current pulse, injected through the micropipette while the cell was at a relatively depolarized potential, generated a regular train of action potentials. In contrast, when the same amplitude pulse was applied on a background of hyperpolarization, it elicited a much stronger response, in the form of a stereotyped all-or-none burst of action potentials. The implication was that rather than decreasing cellular excitability as expected, hyperpolarization actually increased it. Such unusual behavior was at odds with the current understanding of neuronal excitability based on spinal cord studies, and prompted the search for new biophysical mechanisms. The intracellular recordings of Deschenes, Roy, and Steriade (1982) indicated for the first time that the bursting behavior under hyperpolarization was an intrinsic property of thalamic neurons and was not generated by a particular pattern of network activity. But the controversy was only solved definitely with the development of the brain slice technique. Intracellular recordings in thalamic slices by Llinas and Jahnsen (1982) revealed that a single ionic current, a low threshold calcium current termed ‘t’ for transient (Fox, Nowycky, & Tsien, 1987) was responsible for the bursting behavior of thalamic cells under hyperpolarization. The same current had been shown to generate the rhythmic behavior of inferior olivary neurons (Llinas & Yarom, 1981). Those studies are part of a now classic series of works by Llinas and colleagues that revealed that neurons in different nuclei of the mammalian central nervous system exhibit unique sets of electrophysiological properties and culminated in a new view of the functioning of the mammalian central nervous system (Llinas, 1988).

Indeed, the electrophysiology of thalamocortical (TC) relay cells constitutes an excellent example of the role played by the biophysical properties of a class of neurons in defining the functional state of a network. Among many other functions, TC cells relay sensory information from the periphery to the cerebral cortex. During the waking state, TC cells respond to excitatory input with trains of single action potentials whose frequency is related linearly to the amplitude of the input, thus faithfully relaying information to their cortical targets (Steriade, McCormick, & Sejnowski, 1993). Upon the transition to sleep, TC cells respond to the same inputs with all or none stereotyped bursts of action potentials, i.e. they are not modulated in intensity by the amplitude of the inputs and therefore contain little information about the inputs except for their arrival time (with the caveat that TC bursts have a very long refractory period and would not respond to inputs in rapid succession with intervals of less than 50–70 ms). This dramatic transition in firing pattern (from tonic to bursting) plays a key role in the gating function of the thalamus and is one of the most important factors responsible for the deafferentation of the neocortex during slow wave sleep (Steriade, 2000, Steriade and Deschenes, 1984, Steriade and Llinas, 1988, Steriade et al., 1993). This transition in firing pattern is due to the voltage dependence of a single ionic current present in TC cells, the low threshold calcium current It. This current underlies the all or none depolarizing potential called a low threshold spike or LTS (Jahnsen and Llinas, 1984a, Jahnsen and Llinas, 1984b) on top of which rides a burst of 3–4 Na+ spikes characteristic of slow wave sleep. The It current is inactivated by the depolarized resting Vm of the waking state. During the transition to sleep, there is a progressive reduction in the firing rates of neuromodulatory systems (such as cholinergic, adrenergic or serotonergic) of the brainstem and basal forebrain, which project diffusely to thalamus and cortex and maintain leak K+ conductances relatively low (McCormick, 1992, Steriade, 1992, Steriade, 2004b, Steriade et al., 1993). The increase of IKleak during sleep results in widespread neuronal hyperpolarization in TC cells (Hirsch, Fourment, & Marc, 1983) and in most cortical cells, which removes inactivation of the It current and switches the firing mode of TC cells from tonic to bursting.

Section snippets

Electrophysiological cell classes in neocortex

Far from just cellular taxonomy using functional criteria, the classification of cells into electrophysiological categories based on their firing patterns allows the construction of models of local networks that take into account the different dynamics of the individual cells. The biophysical properties of the cell determine the temporal pattern of spike firing in response to sustained activation, thus playing a critical role in the transformation of synaptic input into spike output. The

Intrinsic electrophysiological properties determine firing pattern in response to visual stimulation

A fundamental consequence of the electrophysiological properties described above, is that they determine the output pattern to synaptic input when cells are embedded in a functioning network. An example of the parallel between responses to current injection and visual stimulation is shown in Fig. 2. Three cells, each one belonging to a different electrophysiological cell class, were recorded from the primary visual cortex (area 17) of the anesthetized cat in vivo. The RS cell (top row)

Differences in input–output function among different electrophysiological classes of visual cortex neurons

The input–output function of cortical cells has been extensively studied using direct current injection through the micropipette. It has been shown, as discussed above, that FS cells have higher gain than RS cells (Azouz et al., 1997, Connors and Gutnick, 1990). However, such differences in input–output function had not been established using visual stimuli. In the visual cortex, the input–output function of neurons has been extensively studied using stimulus contrast as the input variable.

Persistent sodium current (INaP)

The persistent sodium current (INaP) was first described in Purkinje cells of the cerebellum (Llinas & Sugimori, 1980) and then in neocortical pyramidal neurons (Connors et al., 1982, Crill, 1996, Stafstrom et al., 1985, Stafstrom et al., 1982). The current is generated by the opening of voltage dependent, TTX sensitive Na+ channels that show no evidence of inactivation. INaP is activated at about −65 mV (below spike threshold), therefore it is activated by the membrane potential trajectory of

Conclusion

The examples illustrated here point to a critical role played by the biophysical properties of single neurons in the dynamics of local circuit networks. However, the detailed knowledge of the physiology or morphology of single cells cannot fully explain the behavior of local circuits, much less their functional roles. The dynamic properties of local circuits are emergent (Mountcastle, 1998). They arise from a combination of single cell properties, synaptic interactions and the actions of

Acknowledgements

Supported by the DoD Multidisciplinary University Research Initiative (MURI) program administered by the Office of Naval Research under grant N00014-01-1-0625, and by NIH-NEI RO1-013984.

References (159)

  • P.S Goldman-Rakic

    Cellular basis of working memory

    Neuron

    (1995)
  • S.T Hammett et al.

    Perceived contrast as a function of adaptation duration

    Vision Research

    (1994)
  • F Helmchen

    Raising the speed limit–Fast Ca(2+) handling in dendritic spines

    Trends in Neuroscience

    (2002)
  • J.C Hirsch et al.

    Sleep-related variations of membrane potential in the lateral geniculate body relay neurons of the cat

    Brain Research

    (1983)
  • Y Kawaguchi et al.

    Neurochemical features and synaptic connections of large physiologically-identified GABAergic cells in the rat frontal cortex

    Neuroscience

    (1998)
  • M Marin-Padilla et al.

    Distribution of the apical dendritic spines of the layer V pyramidal cells of the hamster neocortex

    Brain Research

    (1968)
  • K.A Martin

    Microcircuits in visual cortex

    Current Opinion in Neurobiology

    (2002)
  • D.A McCormick

    Neurotransmitter actions in the thalamus and cerebral cortex and their role in neuromodulation of thalamocortical activity

    Progress in Neurobiology

    (1992)
  • A Agmon et al.

    Correlation between intrinsic firing patterns and thalamocortical synaptic responses of neurons in mouse barrel cortex

    Journal of Neuroscience

    (1992)
  • B Ahmed et al.

    Polyneuronal innervation of spiny stellate neurons in cat visual cortex

    The Journal of Comparative Neurology

    (1994)
  • D Albe-Fessard et al.

    Exploration of certain activities of the motor cortex in cat with microelectrodes; endo-somatic leads

    Journal of Physiology (Paris)

    (1953)
  • D.G Albrecht et al.

    Spatial contrast adaptation characteristics of neurones recorded in the cat's visual cortex

    Journal of Physiology

    (1984)
  • Y Amitai et al.

    Intrinsic physiology and morphology of single neurons in neocortex

    (1995)
  • Y Amitai et al.

    Intrinsic physiology and morphology of single neurons in neocortex

    (1995)
  • Y Amitai et al.

    Regenerative activity in apical dendrites of pyramidal cells in neocortex

    Cerebral Cortex

    (1993)
  • P Andersen et al.

    Inhibitory phasing of neuronal discharge

    Nature

    (1962)
  • R Azouz et al.

    Physiological properties of inhibitory interneurons in cat striate cortex

    Cerebral Cortex

    (1997)
  • L.S Benardo et al.

    Electrophysiology of isolated hippocampal pyramidal dendrites

    Journal of Neuroscience

    (1982)
  • C Blakemore et al.

    Adaptation to spatial stimuli

    Journal of Physiology

    (1969)
  • L.G Brock et al.

    The recording of potentials from motoneurones with an intracellular electrode

    Journal of Physiology

    (1952)
  • J.C Brumberg et al.

    Ionic mechanisms underlying repetitive high-frequency burst firing in supragranular cortical neurons

    Journal of Neuroscience

    (2000)
  • J.M Budd et al.

    Local lateral connectivity of inhibitory clutch cells in layer 4 of cat visual cortex (area 17)

    Experimental Brain Research

    (2001)
  • E.H Buhl et al.

    Physiological properties of anatomically identified axo-axonic cells in the rat hippocampus

    Journal of Neurophysiology

    (1994)
  • W.H Calvin et al.

    Fast and slow pyramidal tract neurons: An intracellular analysis of their contrasting repetitive firing properties in the cat

    Journal of Neurophysiology

    (1976)
  • M Carandini et al.

    A tonic hyperpolarization underlying contrast adaptation in cat visual cortex

    Science

    (1997)
  • M Carandini et al.

    Membrane potential and firing rate in cat primary visual cortex

    Journal of Neuroscience

    (2000)
  • B Cauli et al.

    Molecular and physiological diversity of cortical nonpyramidal cells

    Journal of Neuroscience

    (1997)
  • Y Chagnac-Amitai et al.

    Synchronized excitation and inhibition driven by intrinsically bursting neurons in neocortex

    Journal of Neurophysiology

    (1989)
  • F.S Chance et al.

    Synaptic depression and the temporal response characteristics of V1 cells

    Journal of Neuroscience

    (1998)
  • A Chow et al.

    K(+) channel expression distinguishes subpopulations of parvalbumin- and somatostatin-containing neocortical interneurons

    Journal of Neuroscience

    (1999)
  • C.M Colbert et al.

    Axonal action-potential initiation and Na+ channel densities in the soma and axon initial segment of subicular pyramidal neurons

    Journal of Neuroscience

    (1996)
  • B.W Connors et al.

    Electrophysiological properties of neocortical neurons in vitro

    Journal of Neurophysiology

    (1982)
  • D Contreras et al.

    Voltage-sensitive dye imaging of neocortical spatiotemporal dynamics to afferent activation frequency

    Journal of Neuroscience

    (2001)
  • D Contreras et al.

    Response to contrast of electrophysiologically defined cell classes in primary visual cortex

    Journal of Neuroscience

    (2003)
  • W.E Crill

    Persistent sodium current in mammalian central neurons

    Annual Review of Physiology

    (1996)
  • W.E Crill et al.

    Ionic mechanisms underlying excitation-to-frequency transduction: Studies by voltage clamp methods

    Archives Italiennes de Biologie

    (1984)
  • J DeFelipe

    Neocortical neuronal diversity: Chemical heterogeneity revealed by colocalization studies of classic neurotransmitters, neuropeptides, calcium-binding proteins, and cell surface molecules

    Cerebral Cortex

    (1993)
  • E de la Pena et al.

    Participation of low-threshold calcium spikes in excitatory synaptic transmission in guinea pig medial frontal cortex

    European Journal of Neuroscience

    (2000)
  • R.J Douglas et al.

    A functional microcircuit for cat visual cortex

    Journal of Physiology

    (1991)
  • J Eccles

    The physiology of nerve cells

    (1957)
  • Cited by (0)

    View full text