512.full

doi:10.1152/jn.00126.2006 96:512-521, 2006.J Neurophysiol E. J. Tehovnik, A. S. Tolias, F. Sultan, W. M. Slocum and N. K. Logothetis Electrical Microstimulation Direct and Indirect Activation of Cortical Neurons by You might find this additional info useful... 125 articles, 54 of which can be accessed free at:This article cites http://jn.physiology.org/content/96/2/512.full.html#ref-list-1 16 other HighWire hosted articles, the first 5 are:This article has been cited by [PDF] [Full Text] [Abstract] , March, 2010; 20 (3): 704-719.Cereb. Cortex Marie-Hélène Boudrias, Sang-Pil Lee, Stan Svojanovsky and Paul D. Cheney Wall of Rhesus Macaques Forelimb Muscle Representations and Output Properties of Motor Areas in the Mesial [PDF] [Full Text] [Abstract] , April, 21 2010; 30 (16): 5730-5743.J. Neurosci. Arthur Leblois, Benjamin J. Wendel and David J. Perkel Context-Dependent Behavioral Variability through D1 Receptors Striatal Dopamine Modulates Basal Ganglia Output and Regulates Social [PDF] [Full Text] [Abstract] , July, 2010; 104 (1): 108-118.J Neurophysiol Shai Netser, Shay Ohayon and Yoram Gutfreund Pupil Dilations, and Sensory Priming Multiple Manifestations of Microstimulation in the Optic Tectum: Eye Movements, [PDF] [Full Text] [Abstract] , November, 2010; 104 (5): 2778-2791.J Neurophysiol Daniel K. Freeman, Donald K. Eddington, Joseph F. Rizzo III and Shelley I. Fried Selective Activation of Neuronal Targets With Sinusoidal Electric Stimulation [PDF] [Full Text] [Abstract] , December, 2010; 104 (6): 3462-3475.J Neurophysiol Douglas Crawford Jachin A. Monteon, Alina G. Constantin, Hongying Wang, Julio Martinez-Trujillo and J. Kinematically Normal 3D Gaze Shifts Electrical Stimulation of the Frontal Eye Fields in the Head-Free Macaque Evokes including high resolution figures, can be found at:Updated information and services http://jn.physiology.org/content/96/2/512.full.html can be found at:Journal of Neurophysiologyabout Additional material and information http://www.the-aps.org/publications/jn This infomation is current as of January 8, 2011. American Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at http://www.the-aps.org/. (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2006 by the publishes original articles on the function of the nervous system. It is published 12 times a yearJournal of Neurophysiology o n January 8, 2011 jn.physiology.org D ow nloaded from Invited Review Direct and Indirect Activation of Cortical Neurons by Electrical Microstimulation E. J. Tehovnik,1 A. S. Tolias,2 F. Sultan,3 W. M. Slocum,1 and N. K. Logothetis2 1Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts; 2Max Planck Institute for Biological Cybernetics, Tuebingen; and 3Department of Cognitive Neurology, Hertie-Institute for Clinical Brain Research, University of Tuebingen, Tuebingen, Germany Submitted 5 February 2006; accepted in final form 30 April 2006 Tehovnik, E. J., A. S. Tolias, F. Sultan, W. M. Slocum, and N. K. Logothetis. Direct and indirect activation of cortical neurons by electrical microstimulation. J Neurophysiol 96: 512–521, 2006; doi:10.1152/jn.00126.2006. Electrical microstimulation has been used to elucidate cortical function. This review discusses neuronal excit- ability and effective current spread estimated by using three different methods: 1) single-cell recording, 2) behavioral methods, and 3) functional magnetic resonance imaging (fMRI). The excitability prop- erties of the stimulated elements in neocortex obtained using these methods were found to be comparable. These properties suggested that microstimulation activates the most excitable elements in cortex, that is, by and large the fibers of the pyramidal cells. Effective current spread within neocortex was found to be greater when measured with fMRI compared with measures based on single-cell recording or behavioral methods. The spread of activity based on behavioral methods is in close agreement with the spread based on the direct activation of neurons (as opposed to those activated synaptically). We argue that the greater activation with imaging is attributed to transynaptic spread, which includes subthreshold activation of sites connected to the site of stimulation. The definition of effec- tive current spread therefore depends on the neural event being measured. I N T R O D U C T I O N Electrical stimulation of neural tissue has been in use for over 100 yr (at least since 1870, Fritch and Hitzig) and, although some have argued it is an imprecise technique for understanding the detailed mechanisms underlying different neural computations, microstimulation has actually contributed to important successes both in clinical applications and in uncovering the secrets of how the brain mediates various psychological processes. For example, electrical stimulation has made it possible to restore hearing to deaf patients by delivering microampere pulses by implanted electrodes to different regions of the cochlea (Bierer and Middlebrooks 2002, 2004; Fu 2005; Middlebrooks and Bierer 2002; Snyder et al. 2004). Stimulation of the basal ganglia has been remark- ably effective in restoring motor function to Parkinsonian patients (Dostrovsky and Lozano 2002; Dostrovsky et al. 2000; Limousin et al. 1995; MacKinnon et al. 2005). In the same vein, microstimulation of the visual pathway is currently re- garded as a very promising method for making the blind see again (Bartlett et al. 2005; Bradley et al. 2005; DeYoe et al. 2005; Merabet et al. 2005; Pezaris and Reid 2004; Schmidt et al. 1996; Tehovnik et al. 2005a; Zrenner 2002). Equally important is the contribution of microstimulation in suggesting causal links between brain structures and behavior. Stimulation has been used to study pathways in the brain that subserve reward (Gallistel et al. 1981), as well as pathways involved in locomotion and startle responses (Yeomans and Frankland 1996; Yeomans and Tehovnik 1988). Its use has disclosed topographic maps in primary visual cortex (area V1), the supplementary eye fields, frontal eye fields, and the supe- rior colliculus for the generation of ocular responses (Robinson 1972; Robinson and Fuchs 1969; Scha¨fer 1988; Tehovnik and Lee 1993). Additionally it has led to the elucidation of topo- graphic maps in the motor and supplementary motor areas for the execution of skeletomotor responses (Fritch and Hitzig 1870; Graziano et al. 2002; Penfield and Boldrey 1937; Strick and Preston 1978; Woolsey et al. 1952, 1979). Finally, elec- trical stimulation has been used to study cortical function as it pertains to the sense of vision, hearing, and touch (Britten and van Wezel 1998; DeAngelis et al. 1998; Penfield and Boldrey 1937; Penfield and Perot 1963; Romo et al. 1998; Salzman et al. 1990). Many investigators are currently using electrical micro- stimulation routinely in behaving monkeys to make inferences on how neocortex mediates a range of behaviors, from target selection to avoidance responses (e.g., Cooke et al. 2003; Cutrell and Marrocco 2002; Moore and Armstrong 2003; Moore and Fallah 2001; Opris et al. 2005; Schiller and Tehovnik 2001; Tehovnik et al. 2005a). In all such studies it is important to characterize the neural circuits that are activated during microstimulation both locally about the electrode tip and in projection sites. In this regard, two issues are of paramount importance: the need of accurate estimates of effective current spread and its effects on the excitable elements of the tissue. Our review therefore discusses both issues based on single-cell recording, behavioral methods, and neuroimaging. E F F E C T I V E C U R R E N T S P R E A D U S I N G D I R E C T A C T I V A T I O N O F C O R T I C A L N E U R O N S Spread and excitability properties It is commonly accepted that the sites of direct activation of a neuron with electrical microstimulation are at the initial segment and nodes of Ranvier (Gustaffson and Jankowska 1976; Nowak and Bullier 1998a,b; Porter 1963; Rattay 1999; Swadlow 1992). These zones contain the highest concentra- tions of sodium channels, thereby making them the most excitable segments of a neuron (Catterall 1981; Nowak and Bullier 1998a,b; Waxman and Quick 1978). The amount of Address for reprint requests and other correspondence: E. J. Tehovnik, Massachusetts Institute of Techology, Bldg. 46-6041, Cambridge, MA 02139 (E-mail: tehovnik@mit.edu). J Neurophysiol 96: 512–521, 2006; doi:10.1152/jn.00126.2006. 512 0022-3077/06 $8.00 Copyright © 2006 The American Physiological Society www.jn.org o n January 8, 2011 jn.physiology.org D ow nloaded from current injected through a microelectrode to directly activate a neuron (i.e., cell body or axon) is proportional to the square of the distance between the neuron and the electrode tip. This is expressed as I � Kr2, where I is the current level (in micro- amperes [�A]), r is distance (in millimeters [mm]), and K is the excitability constant (in �A/mm2). This relationship is derived from studies of cortical and corticospinal neurons of rats, cats, and primates (Asanuma et al. 1976; Marcus et al. 1979; Nowak and Bullier 1996; Shinoda et al. 1976, 1979; Stoney et al. 1968); dopaminergic fibers of the medial forebrain bundle in rats (Yeomans et al. 1988); cerebellar and reticulospinal fibers of rats, rabbits, and cats (Akaike et al. 1973; Armstrong et al. 1973b; Hentall et al. 1984a; Jankowska and Roberts 1972; Roberts and Smith 1973); cell bodies of cat spinal motor neurons (Gustaffson and Jankoska 1976); and axons of spinal interneurons of cats (Jankowska and Roberts 1972). In these studies, a single cathodal pulse was used to evoke an anti- dromic extracellular action potential as the stimulating elec- trode was advanced toward and beyond the element being stimulated. The effective current spread from an electrode tip can be expressed as the square root of the current divided by the square root of the excitability constant, i.e., (I/K)1/2. This relationship is illustrated in Fig. 1A for a group of pyramidal tract neurons that were identified antidromically by pontine stimulation in cats (Stoney et al. 1968). To activate a neuron from the cortex, a 0.2-ms cathodal pulse was delivered through a microelectrode (Fig. 1, methods). The amount of current required for the evocation of an action potential 50% of the time defined the current threshold. As the electrode was advanced toward and past the neuron, the current threshold dropped and then increased accordingly (Fig. 1, methods, electrode S: a–c). The rate of change of the current threshold with electrode displacement was used to deduce the excitability constant. For the group of pyramidal tract neurons, the average excitability constant was 1,292 �A/mm2. This constant reflects FIG. 1. Current-spread and excitability properties of pyramidal tract neurons determined using single-cell recordings within motor cortex of the cat. A: radial distance (in millimeters) of direct activation of pyramidal tract neurons using the equation radial distance � (K/I)1/2, where K is the current–distance constant and I is the current used. Curve represents the amount of current required for the antidromic elicitation of an action potential 50% of the time using a single cathodal pulse of 0.2-ms duration. For the 12 cells studied the average K value was 1,292 �A/mm2. Shaded gray area about the curve represents 1 SE, with K values ranging from 1,037 to 1,547 �A/mm2. Data derived from Stoney et al. (1968). B: current threshold normalized to the rheobase current (IRo) is plotted as a function of pulse duration for the direct activation of 6 pyramidal tract neurons. Rheobase current is the current used at the longest pulse duration of 1.0 ms. Each curve represents data from a single neuron. A curve represents the amount of current at a given pulse duration required for the antidromic elicitation of an action potential 50% of the time using a single cathodal pulse. Shaded area represents the range of chronaxies for the neurons stimulated. A chronaxie is the pulse duration at a current level of twice the rheobase current. Data from Asanuma et al. (1976) and Stoney et al. (1968). Method used by Asanuma et al. (1976) and Stoney et al. (1968) to derive the data in A and B is illustrated on the right. Recording electrode (R) and stimulating electrode (S), both situated next to the pyramidal tract cell, are depicted to scale. Exposed microelectrode tips, shown as a black triangle, were constructed to have a diameter of 10 �m and a length of 15 �m. Each dotted circle represents the field of effective stimulation produced by one cathodal pulse that activates the neuron’s initial segment (IS), which is the lowest current threshold site at the cell body before the start of the axon (Gustaffson and Jankowska 1976). Each field of effective stimulation is centered on a different electrode tip (a–c), indicating the path of the stimulating electrode. Electrode tip b is located at the lowest-threshold locus for current as indicated by the smallest field of stimulation. A scale bar is shown. Invited Review 513ACTIVATION BY ELECTRICAL MICROSTIMULATION J Neurophysiol • VOL 96 • AUGUST 2006 • www.jn.org o n January 8, 2011 jn.physiology.org D ow nloaded from the excitability of a neural element 1 mm away from the electrode tip such that an element having a constant of 1,292 �A/mm2 would require a 1,292-�A current to be activated 1 mm away 50% of the time. Increases in the excitability constant have been associated with decreases in the conduction velocity of an axonal element (Hentall et al. 1984a; Jankowska and Roberts 1972; Nowak and Bullier 1996; Roberts and Smith 1973). The conduction velocities of myelinated pyramidal tract neurons range from 3 to 80 m/s with the largest of these neurons exhibiting the highest velocities (Calvin and Sypert 1976; Deschenes et al. 1979; Finlay et al. 1976; Macpherson et al. 1982; Phillips 1956; Takahashi 1965); the conduction velocities of small unmyelinated cortical fibers are �1 m/s (Nowak and Bullier 1996; Swadlow 1985). The excitability constant derived with a 0.2-ms pulse can be as low as 300 �A/mm2 for the largest myelinated cortical neurons and as high to 27,000 �A/mm2 for the smallest unmyelinated cortical neurons (Nowak and Bullier 1996; Stoney et al. 1968). In other words, this constant is inversely related to the size of a neuron’s axon and to whether it is myelinated. Strength–duration functions and estimates of excitability To deduce the excitability of stimulated neurons, current can be traded off against pulse duration to elicit some response (Armstrong et al. 1973a; Asanuma et al. 1976; Bartlett et al. 2005; BeMent and Ranck 1969; Brindley and Lewin 1968; Dobelle and Mladejovsky 1974; Farber et al. 1997; Grumet et al. 2000; Hentall et al. 1984b; Jankowska and Roberts 1972; Li and Bak 1976; Matthews 1977; Nowak and Bullier 1998a; Ronner and Lee 1983; Rushton and Brindley 1978; Sekirnjak et al. 2006; Shizgal et al. 1991; Stoney et al. 1968; Swadlow 1992; Tehovnik and Lee 1993; Tehovnik and Sommer 1997; Tehovnik et al. 2003; Tolias et al. 2005; West and Wolstencroft 1983; Yeomans et al. 1988). This procedure is used to generate a strength–duration function. Normalized strength–duration functions for pyramidal tract neurons are illustrated [Fig. 1B, derived from Asanuma et al. (1976) and Stoney et al. (1968)]. As the pulse duration is increased, the amount of current needed to evoke an action potential 50% of the time diminishes to some asymptotic level; this level is called the rheobase current. The excitability or chronaxie of a stimulated element is expressed as the pulse duration at twice the rheobase current. The range of chronaxies for pyramidal tract neurons is illus- trated in Fig. 1B in gray. These neurons exhibit chronaxies ranging between 0.1 and 0.4 ms. The shorter the chronaxie the more excitable a directly stimulated neural element. Axons have shorter chronaxies than those of cell bodies (axons: 0.03–7 ms; cell bodies: 7–31 ms; Nowak and Bullier 1998a; Ranck 1975), and large, myelinated axons have shorter chronaxies than those of small, nonmyeli- nated axons (large: 0.03–0.7 ms; small: �1.0 ms; Li and Bak 1976; Ranck 1975; West and Wolstencroft 1983). A chronaxie is negatively correlated with the conduction velocity of axons (Nowak and Bullier 1998a; Swadlow 1992; West and Wolsten- croft 1983) and positively correlated with their refractory period (Shizgal et al. 1991). The chronaxie is related to the time constant of the directly stimulated membrane of a neuron (Ranck 1975), which depends on a membrane’s resistance and capacitance (Bostock 1983; Bostock et al. 1983). E F F E C T I V E C U R R E N T S P R E A D U S I N G B E H A V I O R A L M E T H O D S Spread properties Several investigators have used behavioral methods to estimate how far current activates neural tissue mediating behaviors such as eating (Olds 1958), self-stimulation (Fou- riezos and Wise 1984; Milner and Lafarriere 1986; Wise 1972), and lateral head and body movements (Yeomans et al. 1986). These estimates are based on the activation of subcortical fibers. Two groups have studied the current- spread properties of electrical stimulation within neocortex using behavioral methods. Murasugi et al. (1993) studied such properties in extrastriate area MT (middle temporal cortex) and Tehovnik et al. (2004, 2005b) conducted cur- rent-spread studies in striate area V1. Murasugi et al. (1993) stimulated area MT of monkeys with 1-s trains composed of 0.2-ms pulses delivered at 200 Hz to bias a monkey’s discrimination of the direction of dot motion. The motion stimuli were presented in the receptive field of the stimulated neurons, which were tuned to a particular direction of motion. Murasugi et al. found that for the range of currents tested (i.e., 10 to 80 �A), currents �20 �A biased a monkey’s discrimination abilities less well and began to obscure perfor- mance. Thus it can be presumed that �20 �A activates neural tissue confined to roughly one “directional” column in MT. The approximate width of such a column is about 0.2 mm (Albright et al. 1984). The average excitability constant of the activated elements in MT is therefore estimated to be 2,000 �A/mm2 [K � 20 �A/(0.1 mm)2]. Finally, for pulse frequen- cies as high as 500 Hz (using 10-�A current pulses), a monkey’s performance on the discrimination task was never obscured; this suggests that such high frequencies delivered in 1-s trains exhibit effects confined to 0.2 mm of cortical tissue. Tehovnik et al. (2004) found that microstimulation of V1 (with 100-ms trains using 0.2-ms pulses delivered at 200 Hz) systematically delayed the execution of visually guided sac- cades as long as the stimulation was delivered immediately before a monkey generates the saccade (i.e., at the end of the fixation period before the onset of the visual target) and as long as the visual target was punctate (�0.4° of visual angle) and located within the center of the receptive field of the stimulated neurons. No delay effect occurred to targets located outside of the receptive field of the stimulated neurons. It was later found that the size of the visual field affected by the stimulation, called a delay field, varied as a function of the site of stimu- lation within the operculum of V1 and it also varied as a function of current (Tehovnik et al. 2005b). A summary of the data from this study is illustrated in Fig. 2, A and B. As the stimulating electrode was situated further from the foveal representation of V1 the size of the delay field increased. For stimulations of cells having receptive field centers at 2, 3, and 4° from the fovea, the size of the delay field was 0.1