Chapter 5 Electrophysiology

Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

1 of 19

10/05/2009 15:49

Print Close Window

Anthony A. Grace, J. Amiel Rosenkranz, Anthony R. West: Chapter 5. Electrophysiology, in The American Psychiatric Publishing Textbook of

Psychopharmacology, 4th Edition. Edited by Alan F. Schatzberg, Charles B. Nemeroff. Copyright ©2009 American Psychiatric Publishing, Inc.

DOI: 10.1176/appi.books.9781585623860.408404. Printed 5/10/2009 from www.psychiatryonline.com

Textbook of Psychopharmacology >

Chapter 5. Electrophysiology

ELECTROPHYSIOLOGY: INTRODUCTION

Several approaches can be used to analyze the structure and function of the nervous system in health and disease.

Many of these techniques—for example, the biochemical analysis of neurotransmitter and metabolite levels,

anatomical studies of axonal projection sites or neurotransmitter enzymes, and molecular biological studies of

messenger levels and turnover—examine the nervous system at the level of groups or populations of neurons. In

contrast, by its very nature, electrophysiology is oriented toward the physiological analysis of individual neurons. In

this chapter, we describe preparations and techniques that are in a general sense applicable to many systems, with

specific examples from the dopaminergic system to draw on our field of expertise.

The use of electrophysiological techniques for the analysis of neuronal physiology depends on the unique properties of

the neuronal membrane. Like many other cell types, the neuron possesses an electrochemical gradient across its

membrane. The electrochemical gradient itself is a product of two forces: 1) an electrical potential force derived from

the voltage difference between the inside and the outside of the cell and 2) a chemical potential force resulting from

the unequal distribution of ions across the membrane. Cells set up and maintain this electrochemical gradient because

of the selective permeability of their membranes to particular ionic species. Thus, the membrane has a rather high

degree of permeability to ions such as potassium but is relatively impermeable to ions such as sodium and calcium.

In the resting state, cells have a very low internal concentration of sodium and calcium. To achieve this state, the cell

must expend energy (in the form of adenosine triphosphate [ATP] hydrolysis) to extrude sodium from the intracellular

space in exchange for potassium ions. The extrusion of sodium sets up both a chemical gradient (because sodium

attempts to exist in equal concentrations across the membrane) and an electrical gradient (because sodium is

positively charged and is not freely permeable across the membrane; thus, net positive charges are being removed

from inside the cell). To partially counter this electrical gradient, potassium—which is more permeable—flows down

the electrical gradient to become concentrated inside the cell. However, during this process, the cell is also setting up

an opposing chemical gradient, because potassium is achieving higher concentrations within the cell in comparison

with the extracellular environment. When the electrical force drawing potassium into the cell balances the chemical

force of the concentration gradient forcing potassium out of the cell, the membrane is at equilibrium—with high

extracellular sodium concentrations, relatively high intracellular potassium concentrations, and a transmembrane

potential causing the inside of the cell to be negatively charged with respect to its environment.

A typical resting membrane potential for a neuron is rather small, being on the order of –70 mV with respect to the

extracellular fluid. In actuality, potassium itself is not freely permeable. A small electrochemical gradient exists in

most neurons that attempts to force potassium out of the cell and draw the membrane potential to more negative

values. Although the scenario is somewhat more complicated than this (e.g., involving charged proteins and other

ionic species with selective permeabilities), this description approximates how a cell gains an electrochemical gradient

via the energy-dependent extrusion of sodium.

Note that neurons are not the only cells that have transmembrane potentials. In fact, all living cells have an

electrochemical gradient across their membranes that they use for transporting glucose and other essential materials

and accumulating them against a concentration gradient. Such energy-dependent processes are usually coupled to

other gradients from which they derive this energy. For example, a compound may be taken up and concentrated by

linking its transport to sodium, which itself has a large electrochemical gradient in the opposite direction. What makes

the neuron unique is its ability to rapidly change the permeability of its membrane to one or more ion species in a

regenerative manner. This process underlies the generation of an action potential, sets up active propagation of an

action potential down an axon, and triggers the procedure that ultimately results in neurotransmitter release. It also

provides the electrophysiologist with a measure of neuronal activity that can be assayed by recording the electrical

activity generated by the neuron.

The action potential is an active regenerative phenomenon, which means that the events that initiate the action

potential also serve as the force that drives this event to completion. Normally, a given neuron receives information in

the form of synaptic potentials. For example, an axon terminal synapsing on the neuron releases a neurotransmitter,

which binds to the neuron and selectively alters the permeability of its membrane by opening ion channels linked to

its binding site. An ion channel that opens in response to a neurotransmitter is referred to as a ligand-gated channel.

If the neurotransmitter activates a channel that increases the permeability of the membrane to a negatively charged

ion present in high concentrations in the extracellular fluid (e.g., chloride), the influx of chloride down its

electrochemical gradient causes a negative shift in the membrane potential of the cell, thereby increasing the

potential difference across the membrane, or a hyperpolarization of the cell. If activation of this channel causes a

positively charged ion such as sodium to flow down its electrochemical gradient and into the cell, it will cause a briefPrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

2 of 19

10/05/2009 15:49

decrease in the membrane potential (i.e., a depolarization) of the neuron.

Because a change in the membrane potential alters the electrochemical gradient of potassium across the membrane,

potassium ions will flow through their respective channels to restore the membrane to its resting level. Thus, a

neurotransmitter that depolarizes the membrane causes an efflux of potassium ions and a return of the membrane

potential to resting levels. However, if the depolarization is large enough, another type of channel is activated—the

voltage-gated or voltage-dependent sodium channel. In response to a given level of depolarization, this channel

increases its permeability to sodium to allow more of this ion to enter the cell. The result is a further depolarization of

the membrane and consequently an increased activation of this voltage-dependent channel. Because of the

positive-feedback nature of this event, it is referred to as regenerative, because the depolarization augments the very

factor that causes the cell to be depolarized. The membrane potential at which this regenerative process is initiated is

thus the threshold potential for action potential generation, with a hyperpolarization of the cell causing a decrease in

its excitability and a depolarization increasing the likelihood that it will generate an action potential.

The regenerative depolarization of the membrane has limits, however. One limit is the equilibrium potential for

sodium. The equilibrium potential is the membrane potential at which the electrochemical gradient for a particular ion

is zero, with no net flux of the ion across a membrane. This would occur when the membrane potential is sufficiently

positive to oppose the further influx of the positively charged sodium ion across its concentration gradient. Although

this potential is usually about +40 mV in many cells, the action potential does not actually reach this value. Instead,

another voltage-activated channel that is selectively permeable to potassium is activated. The resultant massive

increase in potassium permeability starts to return the membrane potential to its original state, thereby inactivating

the regenerative sodium conductance. The increased potassium permeability is sufficient to drive the membrane

potential negative to the resting potential and toward the equilibrium potential for potassium (i.e., approximately –80

to –90 mV) before the subsequent decrease in voltage-dependent potassium conductance returns the membrane to its

original resting state.

The equilibrium potential of an ion determines the net effect that opening its associated ion channels will have on the

neuron. The equilibrium potential occurs when the membrane is depolarized or hyperpolarized sufficiently to offset

exactly the effects of the concentration gradient on the ion; as a result, no net flux of this ion crosses the membrane.

For example, because of the very high concentration of sodium outside the neuron compared with inside the cell, the

large concentration gradient for sodium across the membrane attempts to force sodium into the neuron. Therefore, to

oppose this concentration gradient, the membrane potential of the neuron would have to be highly positive to provide

an electrical gradient of equivalent force. This occurs at approximately +40 mV for sodium. However, potassium’s

equilibrium potential is about 10–20 mV more negative than its resting potential, which is partly the result of ATP

hydrolysis that exchanges extruded sodium for potassium. As a result, increasing potassium permeability causes a

hyperpolarization of the neuron because of an efflux of potassium down its concentration gradient.

Chloride is another common ionic species. This ion is negatively charged and therefore has an electrical gradient that

would act against its entering the cell. However, the concentration of chloride is so much higher in the extracellular

fluid that the chemical gradient predominates. As a result, opening chloride ion channels causes chloride to flow into

the cell, hyperpolarizing the membrane (Figure 5–1). In fact, the opening of chloride ion channels is the mechanism

through which the primary inhibitory neurotransmitter in the brain (i.e., -aminobutyric acid [GABA]) decreases

neuronal activity.

FIGURE 5–1. Determining the ionic nature of a synaptic event.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

3 of 19

10/05/2009 15:49Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

4 of 19

10/05/2009 15:49

At least three techniques can be used to determine the ionic species that mediates a synaptic response: determining the reversal

potential of the ion, reversing the membrane potential deflection produced by changing the concentration gradient of the ion

across the membrane, and determining the reversal potential (or blocking the synaptic response) after applying a specific ion

channel blocker. In this figure, three techniques are used to illustrate the involvement of a chloride ion conductance increase

evoked in dopamine-containing neurons by stimulation of the striatonigral -aminobutyric acid (GABA)ergic projection. (A) The

reversal potential of a response may be determined by examining the amplitude of the response as the membrane potential of the

neuron is varied. In this example, we superimposed several responses of the neuron evoked at increasingly hyperpolarized

membrane potentials (top traces), with the membrane potential altered by injecting current through the electrode and into the

neuron (bottom traces = current injection). (A1) A synaptic response in the form of an inhibitory postsynaptic potential (IPSP) is

evoked in a dopamine neuron by stimulating the GABAergic striatonigral pathway (arrow). When increasing amplitudes of

hyperpolarizing current (lower traces) are injected into the neuron through the electrode, a progressive hyperpolarization of the

membrane occurs (top traces). As the membrane is made more negative, the IPSP diminishes in amplitude, eventually being

replaced by a depolarizing response. (A2) Plotting the amplitude of the evoked response (y-axis) against the membrane potential

at which it was evoked (x-axis) illustrates how the synaptic response changes with membrane potential. The membrane voltage at

which the synaptic response is equal to zero (i.e., ~69.2 mV in this case) is the reversal potential of the ion mediating the

synaptic response (i.e., the potential at which the electrochemical forces working on the ion are zero). Therefore, there is no net

flux of ions that cross the membrane. At more negative membrane potentials, the flow of the ion is reversed, causing the chloride

ion (in this case) to exit the cell and result in a depolarization of the membrane. (B) The flow of an ion across a membrane may

also be altered by changing the concentration gradient of the ion across the membrane. Normally, chloride ions flow from the

outside of the neuron (where they are present at a higher concentration) to the inside of the neuron (where their concentration is

lower), causing the membrane potential to become more negative. In this case, the concentration of chloride ions across the

membrane of the dopamine neuron is reversed by using potassium chloride as the electrolyte in the intracellular recording

electrode. (B1) Soon after the neuron is impaled with the potassium chloride–containing electrode, stimulation of the striatonigral

pathway (arrow) evokes an IPSP (bottom trace). However, as the recording is maintained, chloride is diffusing from the electrode

into the neuron, causing the electrochemical gradient to decrease progressively over time. As a result, each subsequent

stimulation pulse evokes a smaller IPSP, eventually causing the IPSP to reverse to a depolarization (top trace). The depolarization

is caused by an efflux of chloride ions out of the neuron and down its new electrochemical gradient. This has caused the reversal

potential of the chloride-mediated response to change from a potential that was negative to the resting potential to one that is

now positive to the resting potential. (B2) After injecting chloride ions into the neuron, spontaneously occurring IPSPs that were

not readily observed in the control case are now readily seen as reversed IPSPs (i.e., depolarizations) occurring in this dopamine

neuron recorded in vivo. (C) Another means for determining the ionic conductance involved in a response is by using a specific ion

channel blocker. This can be done in two ways: by using the drug to block an evoked response or (as shown in this example) by

examining the effects of administering the drug on the neuron to determine whether the cell is receiving synaptic events that alter

the conductance of the membrane to this ion. To do this, the current–voltage relationship of the cell is first established. This is

done by injecting hyperpolarizing current pulses into the neuron (x-axis) and recording the membrane potential that is present

during the current injection (y-axis). These values are then plotted on the graph (filled circles), with the resting membrane

potential being the membrane potential at which no current is being injected into the neuron (y-intercept). The slope of the

resultant regression line (solid line) is equal to the input resistance of the neuron (Rinput = 36 megohms). After administration of

the chloride ion channel blocker picrotoxin (open boxes), a new current–voltage relationship is established in a similar manner.

Picrotoxin caused a depolarization of the membrane (y-intercept of dashed line is more positive) and an increase in the neuron

input resistance (the slope of the dashed line is larger). The intersection of the membrane current–voltage plots obtained before

and after picrotoxin administration is then calculated. By definition, this point of intersection (i.e., ~75 mV) is the reversal

potential of the response to picrotoxin, because a neuron at this membrane potential would show no net change in membrane

potential on drug administration.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

5 of 19

10/05/2009 15:49

Source. Adapted from Grace AA, Bunney BS: “Opposing Effects of Striatonigral Feedback Pathways on Midbrain Dopamine Cell

Activity.” Brain Research 333:271–284, 1985. Copyright 1985, Elsevier. Used with permission.

Neurons within the vertebrate nervous system have additional conductances that provide them with unique functions.

One of these conductances is the voltage-gated calcium conductance. Like sodium, calcium exists in higher

concentrations outside the neuron compared with inside the cell. However, the gradient is even more extreme than it

is for sodium. Even though much less calcium than sodium is present in the extracellular fluid, the equilibrium

potential for calcium is almost +240 mV because of the extremely low intracellular concentration of this ion. The

neuron maintains this low intracellular concentration so as to use this ion for specialized purposes. Thus, calcium

influx causes neurotransmitter release, activates calcium-gated ion channels, and triggers second-messenger systems

(e.g., calcium-regulated protein kinase). Calcium channels, like their sodium counterparts, are also voltage gated and

cause calcium influx into the neuron during the action potential. Furthermore, calcium can influence the excitability of

the cell by activating the calcium-activated potassium current, which then causes a large membrane hyperpolarization

after the spike, known as an afterhyperpolarization, that delays the occurrence of a subsequent spike in that neuron.

After entering the neuron, calcium is rapidly sequestered into intracellular organelles to terminate its action and reset

the neuron before the next event. Therefore, calcium can alter the physiological activity and the biochemical

properties of the neuron it affects (Llinás 1988).

In addition to the role of ion channels in setting the resting membrane potential, generating the action potential, and

repolarizing the membrane potential, they can have a more subtle effect on aspects of electrophysiological function.

For example, by regulating properties of the neuronal membrane, such as the amplitude and speed of the neuronal

voltage response to inputs, ion channels can regulate neuronal responses to synaptic inputs. By modulation of these

parameters, ion channels influence the integration of multiple synapses (temporal and spatial integration) and

ultimately regulate the ability of synaptic inputs to depolarize membrane potentials and drive the generation of action

potentials. Furthermore, neurotransmitters such as acetylcholine and monoamines can influence these same

parameters through modulation of ion channels.

ELECTROPHYSIOLOGICAL TECHNIQUES

Through use of a broad range of electrophysiological techniques to assess information about a neuron, such as those

described earlier in this chapter, experiments can be designed to investigate differences in the physiological

properties of neurons of interest, the interaction between neurotransmitter systems involved in behavioral or

pathological conditions, and the mode of action of pathomimetic or psychotherapeutic agents. In attempting to gain

such information, it is important to note that there is no “best” technique. Each approach has its relative strengths

and weaknesses, and only through integrating information gained at these various levels will a more complete

comprehension of neuronal function be achieved.

Numerous parameters of neuronal activity can be assessed electrophysiologically. These parameters can be selectively

assessed depending on the method of recording used. Six recording methods are reviewed here:

electroencephalographic (EEG) recordings, field potential recordings, single-unit extracellular recordings, intracellular

recordings, patch clamp recordings, and whole-cell recordings. Which parameter is measured is essentially a function

of the type of electrode used.

Electroencephalographic Recordings

In recording EEGs, the desired signal is very small in amplitude; thus, a large electrode that sums activity over large

regions of the brain surface is used. Although this technique is less invasive than others, the information it yields is

comparatively narrow, in that a large array of neurons must be simultaneously activated for the potentials to be

recorded at the scalp. As a result, stimulus presentation and EEG averaging are typically required to separate the

signal desired from the background noise.

Field Potential Recordings

The next level of analysis is the recording of field potentials. This method uses a recording electrode with a smaller tip

and a higher resistance than that for EEG electrodes, and the electrophysiological measures are confined to a small

population of neurons surrounding the electrode tip. This technique still depends on the simultaneous activation of a

number of cells; however, because the electrode is inserted into the brain, the cells do not have to be at the surface of

the skull as in the EEG recordings. Furthermore, the activation can consist of stimulation of an afferent pathway.

Nonetheless, the array of neurons sampled must have a common orientation for the massed activity to be measurable.

Therefore, such measures are typically restricted to cortical structures such as the neocortex and hippocampus. With

this method, the current resulting from the parallel activation of excitatory and inhibitory afferents can be measured

as well as the electrophysiological response of a population of neurons to such stimulation.

On the other hand, even in structures without a parallel orientation of neurons, field recordings have been useful in

evaluating changes in afferent synaptic strength. For example, field recordings have been particularly effective in

examining phenomena manifesting synaptic plasticity, such as long-term potentiation. By measuring the fields

generated by stimulating afferent inputs, one can evaluate how effective synaptic pathways are in driving a region and

how this effectiveness can be modified by experience. Therefore, learning paradigms, behavioral conditioning, and

drug sensitization have all been shown to alter evoked field potentials within functional pathways in the brain (e.g.,Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

6 of 19

10/05/2009 15:49

Goto and Grace 2005b).

Single-Unit Extracellular Recordings

The next level of recording involves examining the electrophysiology of individual neurons with extracellular

single-unit recording techniques. This technique requires that an electrode be placed in close proximity to a single

neuron to record its spike discharge. An electrode with a smaller tip and a higher resistance than those used for

recording field potentials results in the sampling of a smaller volume of tissue (i.e., the somata of individual neurons).

Because the region sampled by the electrode is small, the signal is larger in amplitude and the background noise is

less. This allows easy recording of the spontaneous spike discharge of a single neuron within the brain of a living (but

typically anesthetized) animal. The cells examined are located through use of an atlas and a stereotaxic apparatus.

The stereotaxic apparatus holds the head of the animal in a precise orientation so that a brain atlas may be used to

place the recording electrodes accurately within the region of the brain desired. Furthermore, if a dye is dissolved in

the electrolyte within the recording pipette, the dye may be ejected into the recording site for subsequent histological

verification of the region recorded.

Because the recording electrode is placed near the outside surface of the neuron, there is less of a concern that the

activity recorded is a result of damage to the neuron itself, as may be the case with intracellular recording techniques.

Furthermore, many neurons may be sampled in a given animal. However, as a consequence, the amount of information

that can be obtained from a neuron is limited. Typically, the research is relegated to recording information related to

action potential firing (e.g., the firing rate of the neuron, its pattern of spike discharge, and how these states of

activity may be affected by stimulation of an afferent pathway or administration of a drug [Figure 5–2]). Nonetheless,

when combined with the appropriate pharmacological techniques, extracellular recording has yielded a substantial

amount of valuable information related to drug action or neuronal interconnections of physiologically important

neuronal types. For example, by using a series of coordinated pharmacological and physiological techniques, we were

able to define a unique extracellular waveform as that associated with the discharge of a dopamine-containing neuron

(Bunney et al. 1973; Grace and Bunney 1983). This provided the basis for studies that yielded information defining the

mode of action of antipsychotic drugs (Bunney and Grace 1978; Grace 1992).

FIGURE 5–2. Detection of changes in firing rates and pattern of spike discharge by extracellular recording

measurements.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

7 of 19

10/05/2009 15:49

Extracellular recording techniques are an effective means of assessing the effects of afferent pathway stimulation or drug

administration on neuron activity. On the other hand, the measurements that can be made are typically restricted to changes in

firing rates or in the pattern of spike discharge. (A) This firing rate histogram illustrates the response of a substantia nigra–zona

reticulata neuron to stimulation of the -aminobutyric acid (GABA)ergic striatonigral pathway. A common method for illustrating

how a manipulation affects the firing rate of a neuron is by constructing a firing rate histogram. This is typically done by using

some type of electronic discriminator and counter to count the number of spikes that a cell fires in a given time. In this example,

the counter counts spikes over a 10-second interval and converts this number to a voltage, which is then plotted on a chart

recorder. The counter then resets to zero and begins counting spikes over the next 10-second interval. Therefore, in this firing

rate histogram, the height of each vertical line is proportional to the number of spikes that the cell fires during each 10-second

interval, with the calibration bar on the left showing the equivalent firing frequency in spikes per second. During the period at

which the striatonigral pathway is stimulated (horizontal bars above trace marked “STIM”), the cell is inhibited, as reflected by the

decrease in the height of the vertical lines. When the stimulation is terminated, a rebound activation of cell firing is observed. (B)

In this figure, a similar histogram is used to illustrate the effects of a drug on the firing of a neuron. (B1) This figure shows the

well-known inhibition of dopamine neuron firing rate on administration of the dopamine agonist apomorphine (APO). Each of the

filled arrows represents the intravenous administration of a dose of APO. After the cell is completely inhibited, the specificity of the

response is tested by examining the ability of the dopamine antagonist haloperidol (HAL [open arrow]) to reverse this response.

Typically, drug sensitivity is determined by administering the drug in a dose–response fashion. This is done by giving an initial

drug dose that is subthreshold for altering the firing rate of the cell. The first dose is then repeated, with each subsequent dose

given being twice that of the previous dose. This is continued until a plateau response is achieved (in this case, a complete

inhibition of cell discharge). (B2) The drug is administered in a dose–response manner to facilitate the plotting of a cumulative

dose–response curve, with drug doses plotted on a logarithmic scale (i.e., a log dose–response curve). To compare the potency of

two drugs or the sensitivity of two cells to the same drug, a point on the curve is chosen during which the fastest rate of change of

the response is obtained. The point usually chosen is that at which the drug dose administered causes 50% of the maximal change

obtained (i.e., the ED50). As is shown in this example, the dopamine neurons recorded after a partial dopamine depletion (dashed

line) are substantially more sensitive to inhibition by APO than the dopamine neurons recorded in control (solid line) rats. (C) In

addition to determining the firing rate of a neuron, extracellular recording techniques may be used to assess the effects of drugs

on the pattern of spike discharge. This is typically done by plotting an interspike interval histogram. In this paradigm, a computer

is connected to a spike discriminator, and a train of about 500 spikes is analyzed. The computer is used to time the delay between

subsequent spikes in the train (i.e., the time interval between spikes) and plots this in the form of a histogram, in which the x-axis

represents time between subsequent spikes and the y-axis shows the number of interspike intervals that had a specific delay (bin

= range of time; e.g., for 1-msec bins, all intervals between 200.0 and 200.99 msec). (C1) The cell is firing irregularly (as shown

by the primarily normal distribution of intervals around 200 msec), with some spikes occurring after longer-than-average delays

(i.e., bins greater than 400 msec, probably caused by spontaneous inhibitory postsynaptic potentials [IPSPs] delaying spike

occurrence). (C2) In contrast, this cell is firing in bursts, which consist of a series of 3–10 spikes with comparatively short

interspike intervals (i.e., approximately 70 msec) separated by long delays between bursts (i.e., events occurring at greater than

150-msec intervals). The computer determined that in this case, the cell was discharging 79% of its spikes in bursts, compared

with 0% in (C1).

Source. (A) Adapted from Grace AA, Bunney BS: “Opposing Effects of Striatonigral Feedback Pathways on Midbrain Dopamine Cell

Activity.” Brain Research 333:271–284, 1985. Copyright 1985, Elsevier. Used with permission. (B) Adapted from Pucak ML, Grace

AA: “Partial Dopamine Depletions Result in an Enhanced Sensitivity of Residual Dopamine Neurons to Apomorphine.” SynapsePrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

8 of 19

10/05/2009 15:49

9:144–155, 1991. Copyright 1991, Wiley. Used with permission. (C) Adapted from Grace AA, Bunney BS: “The Control of Firing

Pattern in Nigral Dopamine Neurons: Single Spike Firing.” Journal of Neuroscience 4:2866–2876, 1984a. Copyright 1984, Society

for Neuroscience. Used with permission.

Extracellular recordings from neurons measure the current flow generated around a neuron as it generates spikes. For

this reason, extracellular action potentials generally are composed of two components: a positive-going component

followed by a negative-going component. The positive-going component is a reflection of the ion flux across the

neuronal membrane surrounding the electrode that occurs during the depolarizing phase of the action potential, with

the negative phase reflecting the repolarization. Because the extracellular recording electrode is measuring current

across the membrane occurring in concert with changes in intracellular membrane potential and because current is

defined in terms of the first derivative (i.e., rate of change) of voltage, the extracellularly recorded action potential (or

spike) waveform is typically a first derivative of the action potential voltage with respect to time (Terzuolo and Araki

1961). This phenomenon underlies the biphasic nature of the extracellularly recorded event (Figure 5–3).

Furthermore, the recorded spike is largest when the recording electrode is placed near the active site of spike

generation because the current density is greatest (and thus the voltage drop induced across the electrode largest) at

this site.

FIGURE 5–3. Relationship between action potentials recorded intracellularly and those recorded extracellularly from

dopamine-containing neurons.

(A) During intracellular recordings, an action potential is initiated from a negative resting membrane potential (e.g., ~55 mV),

reaches a peak membrane potential (solid arrow), and is followed by a repolarization of the membrane and usually an

afterhyperpolarization. An inflection in the rising phase of the spike (open arrow) is often observed. This reflects the delay

between the initial segment spike that initiates the action potential (occurring prior to the open arrow) and the somatodendriticPrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

9 of 19

10/05/2009 15:49

action potential that it triggers (occurring after the open arrow). (B) A computer was used to differentiate the membrane voltage

deflection occurring in the action potential in (A) with respect to time, resulting in a pattern that shows the rate of change of

membrane voltage. Note that the inflection is exaggerated (open arrow), and the peak of the action potential crosses zero (solid

arrow), because at the peak of a spike, the rate of change reaches zero before reversing to a negative direction. (C) A trace

showing a typical action potential in a dopamine neuron recorded extracellularly. The extracellular action potential resembles the

differentiated intracellular action potential in (B). This is because the extracellular electrode is actually measuring the current

crossing the membrane during the action potential and is therefore, by definition, equivalent to the absolute value of the first

derivative of the voltage trace in (A). The amplitude of the extracellular spike is indicated in volts, because the parameter

measured is actually the voltage drop produced across the electrode tip by the current flux and is therefore much smaller than the

actual membrane voltage change that occurs in (A).

Source. Adapted from Grace AA, Bunney BS: “Intracellular and Extracellular Electrophysiology of Nigral Dopaminergic Neurons, I:

Identification and Characterization.” Neuroscience 10:301–315, 1983. Copyright 1983, International Brain Research Organization.

Used with permission.

Intracellular Recordings

With intracellular recording, an electrode with a much smaller tip and a much higher electrical resistance than that

used with extracellular recording is inserted into the membrane of the neuron. Although the tip of the electrode is

smaller, the signal measured is much larger than that with extracellular recording, because it measures the potential

difference across the membrane directly rather than relying on transmembrane current density changes outside the

neuron. As a result, one can measure electrical activity occurring within the neuron that would be nearly impossible to

measure extracellularly, such as spontaneously occurring or evoked (via afferent pathway stimulation) electrical

potentials generated by neurotransmitter release (or postsynaptic potentials). Furthermore, because the membrane

potential of the neuron may be altered by injecting depolarizing or hyperpolarizing current into the cell through the

electrode, the equilibrium potential (also known as the reversal potential) of the response may be determined. In

addition, the overall conductance of the membrane may be measured by injecting known levels of current and

measuring the membrane voltage deflection produced. By applying this information using Ohm’s law, the input

resistance of the cell can be determined. This could be important in assessing drug effects. A drug could increase the

input resistance of the membrane, making it more responsive to current generated by afferent synapses, without

changing the membrane potential of the neuron. Indeed, such a condition has been proposed to underlie the

mechanism through which norepinephrine exerts a “modulatory” action—that is, increasing the amplitude of the

response of a neuron to a stimulus without affecting its basal firing rate (which has also been described as increasing

its “signal-to-noise” ratio; Freedman et al. 1977; Woodward et al. 1979).

The intracellular recording electrode can also be used to inject specific substances into the neuron. For example,

second messengers or calcium chelators can be introduced into the neuron to examine how they regulate the

excitability and modulate the baseline activity of the neuron or how they affect the neuron’s response to drugs (Figure

5–4). Furthermore, by injecting the neuron with a fluorescent dye or enzymatic marker, the morphology of the specific

cell impaled may be recovered and examined (Figure 5–5). This technique can be combined with

immunocytochemistry to examine the neurotransmitter synthesized by the cell under study (e.g., Grace and Onn

1989).

FIGURE 5–4. Effects of intracellular manipulations of cGMP levels on basal activity of striatal medium spiny neurons.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

10 of 19

10/05/2009 15:49

Intracellular application of selective pharmacological agents enables the investigator to examine the direct effects of these agents

on the membrane activity of single neurons as well as to manipulate intracellular second-messenger systems. This figure

demonstrates that manipulation of intracellular cyclic guanosine monophosphate (cGMP) levels potently and specifically modulates

the membrane activity of striatal medium spiny neurons in a manner that cannot be achieved by extracellular application of drugs.

Striatal neurons were recorded after intracellular application (~5 minutes) of either A) vehicle (control), a 0.5% solution of

dimethylsulfoxide (DMSO); B) the drug 1H-[1,2,4]oxadiazole[4,3-a]quinoxalin-1-one (ODQ), which blocks cGMP synthesis by

inhibiting the synthetic enzyme guanylyl cyclase; C) ODQ plus cGMP; or D) the drug zaprinast, which inhibits phosphodiesterase

enzymes responsible for degrading cGMP. (A) Left: After vehicle injection, striatal neurons exhibited typical rapid spontaneous

shifts in steady-state membrane potential and irregular spontaneous spike discharge. Right: Time interval plots of membrane

potential activity recorded from control neurons demonstrated bimodal membrane potential distributions indicative of bistable

membrane activity. (B) Left: Striatal neurons recorded after ODQ injection exhibited significantly lower-amplitude depolarizing

events compared with vehicle-injected controls and rarely fired action potentials. Right: The depolarized portion of the membrane

potential distribution of neurons recorded after ODQ injection was typically shifted leftward (i.e., hyperpolarized) compared with

controls. (C) Left: Striatal neurons recorded after ODQ and cGMP coinjection rarely fired action potentials but exhibited

high-amplitude depolarizing events with extraordinarily long durations. Right: The membrane potential distribution of neurons

recorded after ODQ and cGMP coinjection was similar to that of controls, indicating that cGMP partially reversed some of the

effects of ODQ. (D) Left: Striatal neurons recorded after intracellular injection of zaprinast exhibited high-amplitude depolarizing

events with extraordinarily long durations. Additionally, all of the cells fired action potentials at relatively high rates (0.4–2.2 Hz).

Right: The membrane potential distribution of these neurons was typically shifted rightward (i.e., depolarized) compared with

controls. Because zaprinast blocks the degradation of endogenous cGMP, we can conclude that basal levels of cGMP depolarize the

membrane potential of striatal neurons and facilitate spontaneous postsynaptic potentials. Arrows indicate the membrane potentialPrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

11 of 19

10/05/2009 15:49

at its maximal depolarized and hyperpolarized levels.

Source. Adapted from West AR, Grace AA: “The Nitric Oxide–Guanylyl Cyclase Signaling Pathway Modulates Membrane Activity

States and Electrophysiological Properties of Striatal Medium Spiny Neurons Recorded In Vivo.” Journal of Neuroscience

24:1924–1935, 2004. Copyright 2004, Society for Neuroscience. Used with permission.

FIGURE 5–5. Intracellular staining of neuron recorded intracellularly.

During intracellular recordings, the recording pipette is filled with an electrolyte to enable the transmission of membrane voltage

deflections to the preamplifier. The electrode may also be filled with substances, such as a morphological stain, for injection into

the impaled neuron. In this example, the electrode was filled with the highly fluorescent dye Lucifer yellow. Because this dye has a

negative charge at neutral pH, it may be ejected from the electrode by applying a negative current across the electrode, with the

result that the Lucifer yellow carries the negative current flow from the electrode and into the neuron. Because this dye diffuses

rapidly in water, it quickly fills the entire neuron impaled. The tissue is then fixed in a formaldehyde compound, the lipids clarified

by dehydration-defatting or by using dimethylsulfoxide (Grace and Llinás 1985), and the tissue examined under a fluorescence

microscope. In this case, a brightly fluorescing pyramidal neuron in layer 3 of the neocortex of a guinea pig is recovered.

Inserting an electrode into the membrane of a cell to measure transmembrane voltage and manipulating its membrane

by injecting current are commonly known as current clamp, because the amount and direction of ionic current crossing

the membrane of the cell can be controlled by the experimenter, such as when determining the reversal potential of a

response. Another technique that is effective in neurophysiological research is the use of voltage clamping. With

voltage clamping, the membrane potential of the neuron is maintained at a set voltage level by injecting current into

the cell. This is achieved by rapid feedback electronics that adjusts the current injected to accurately offset any factors

that may act to change this potential. Thus, when the neuron is exposed to a drug that opens ion channels, the effect

of the ionic influx is precisely counterbalanced by altering the current injected into the neuron by the voltage clamp

device. The amount of additional current that must be injected into the neuron to maintain the membrane potential at

its set point is therefore the inverse of the transmembrane current generated in response to the drug. By using

specific ion channel blockers or by altering the extracellular ionic environment of the neuron, the precise ionic

mechanism and conductance changes induced in a neuron by a drug or neurotransmitter may be determined.

Patch Clamp Recordings

A final level of analysis to be discussed is one directed at assessing the response of individual ionic channels in the

membrane of a neuron. Actually, this technique may be described more as a type of high-resolution extracellular

recording. In this method, a glass pipette with a comparatively large tip is drawn, and the tip is fire-polished until it is

very smooth. The electrode tip is then placed against the membrane surface of a neuron under visual control.

Typically, the neuronal membrane is first cleaned of debris with a jet of fluid to permit a tight seal between the

electrode and the membrane. A small suction is then applied to the pipette to tighten its seal with the membrane.

Because such an attachment provides a high-resistance junction with the membrane, the minute transmembranePrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

12 of 19

10/05/2009 15:49

currents that are generated as a result of the opening and closing of individual ionic channels may be monitored. The

biophysical characteristics (e.g., open time, inactivation rate) of specific ion channels and how they may be modified

by drugs applied via the pipette lumen or to other regions of the neuron being studied can be determined.

Whole-Cell Recordings

A technology widely used in neuropharmacology research is whole-cell recordings. This technique is a modification of

the patch clamp technique. The membrane patch underlying the electrode is ruptured, typically by applying a small

suction through the pipette, which results in the interior of the electrode becoming continuous with the intracellular

fluid. The combination of a tight giga-ohm seal around the electrode–cell membrane junction and the very low

resistance of the electrode–intracellular patch has several major advantages: the recordings are very low noise, and

the low resistance of the electrodes can minimize errors that may arise during voltage clamp recordings or distort the

recorded voltage response to a current input.

When combined with infrared video microscopy techniques, which allow both the large electrode tip and the cell to

which it is to be attached to be visualized (Figure 5–6), recordings can be made from specific cell types or from cells

labeled with a retrogradely transported fluorescent dye. Although such a preparation has many unique and powerful

advantages, this low-resistance junction is also subject to the introduction of artifacts. This is particularly true in

cases in which the response to be measured is mediated by diffusible second messengers: the large bore of the

attached electrode has been reported to dialyze intracellular constituents from the cell into the electrode. When this

process happens, the experimenter often observes a rundown of the response, in which the current gradually

decreases with time as a result of loss of the intracellular milieu. To test for this possibility, investigators often rely on

a “perforated patch” technique, in which the patch pipette is filled with the ionophore nystatin, gramicidin, or others.

When a patch pipette of this type is attached to the cell surface, these channel ionophores are incorporated into the

membrane section adjacent to the bore of the electrode. As a result, a low-resistance access to the intracellular space

is obtained without the need for rupturing the membrane.

FIGURE 5–6. Patch clamp electrophysiology and calcium imaging.

By combining patch clamping with injection of selective dyes, the dynamics of calcium can be imaged in real time within isolated

neurons. (A) Using infrared differential interference contrast (IR-DIC) microscopy, the image of a patch pipette can be observed

attached to a neuron during a whole-cell electrophysiology experiment. Note the relatively large size of the pipette tip (coming

from the left side of the image). Calibration bar = 2 m. (B) After filling the neuron with a calcium-sensitive dye, bis-Fura 2, the

live neuron can be imaged with a fluorescence microscope. The dye takes about 10 minutes to fill the neuron after rupturing the

patch membrane. Calibration bar = 10 m. (C) A unique property of the dye bis-Fura 2 is that it changes its fluorescence

properties as it binds calcium. This can be observed by the changes in the fluorescence signal in response to a single (top) or five

(bottom) action potentials. The fluorescence traces correspond to two regions, one close to the cell body (red box and red trace)

and one farther out in the apical dendrite (orange box, orange trace). In this way, one can observe changes in calcium dynamics

and how they correspond to activity states within single neurons.

When penetrating at the level of the soma, the somatic current clamp and voltage clamp recordings are capable of

measuring the voltage and conductance changes that occur at the soma or proximal dendrites. As a result, the

complex membrane dynamics and synaptic events that occur along the dendritic tree cannot be resolved by somatic

penetration in neurons with a complex morphology. One powerful application of patch recording techniques is

targeted recordings of dendrites and presynaptic terminals, structures previously not accessible to direct

electrophysiological measurement. Recordings at dendritic sites have demonstrated dendritic membrane properties

that often are quite distinct from properties determined from somatic recordings. From these findings, it can be

inferred that dendrites are not just extensions of the soma but also play an active role in the propagation of

postsynaptic potentials from the synapse to the soma.

However, many structures, such as finer dendrites and smaller axonals, are still resistant to direct electrophysiological

measurement with patch electrodes. Another technique that allows for examination of minute structures is the use of

fluorescent dyes. Two basic classes of dyes are commonly used: 1) dyes sensitive to changes in ion concentration,

which indirectly reflect the membrane potential, and 2) dyes sensitive to voltage changes, with fluorescence directly

responsive to changes of membrane potential. Because these dyes readily diffuse throughout the neuron, real-time

imaging of changes in dye fluorescence can be performed in live cells (see Figure 5–6). Although the temporal

resolutions and signal-to-noise ratios of imaging technologies do not rival those achievable with direct electrical

recordings, imaging allows examination of minute structures and wide spatial regions. Major drawbacks of this

technique include the toxicity of some dyes and the phototoxicity induced by light sources used to excite the dyes.

More recently, this technique has been applied in vivo, providing new avenues for study of neuronal function.

PREPARATIONS USED IN ELECTROPHYSIOLOGICAL RESEARCH

As with the various types of recordings that can be done, several preparations also can be used in this analysis. No

single preparation is “best”; instead, each has specific advantages and shortcomings. A more complete picture of the

functioning of a system can be gained by taking advantage of the unique perspective provided by each preparation and

designing the experiments accordingly. Except for the first category listed below, all preparations pertain to the

mammalian vertebrate.

Simpler Nervous Systems—Invertebrates and Lower Vertebrate Preparations

We include a reference to simpler nervous systems for completeness; more comprehensive reviews of the use of

nonmammalian model systems can be found elsewhere (Kandel 1978). However, depending on the application, use of

these preparations may yield varying degrees of relevance. With respect to the use of phylogenetically lower species

as models for psychopharmacological studies in humans, much of the data related to anatomy, cellular physiology, and

behavior would be of limited value. The nervous system of these organisms is substantially different from those of

vertebrates and humans, even at the single neuronal level. As a result, information derived from these systems is

likely to be substantially less applicable to behavioral control in the mammalian class. Nonetheless, several unique

advantages are associated with the study of the nervous systems of these organisms: the nervous system is more

accessible, the small number of neurons allows for simple and replicable identification of specific neurons, the largePrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

13 of 19

10/05/2009 15:49

neuronal size enables more stable impalement and more complex procedures, and so on. Furthermore, information

about the study of some second-messenger systems or receptor transduction mechanisms appears to be more directly

transferable to the vertebrate. Thus, it appears that nature is more likely to conserve the most basic functional units of

neurotransmitter actions throughout phylogeny, with decreasing levels of homology as the functional units are

assembled into more complex systems of neurons and networks.

In Vivo Electrophysiological Recordings

Protocols that use the in vivo preparation focus on the living, intact, anesthetized animal as the subject of the study.

Recording the activity of neurons in the intact animal has numerous advantages over studies of isolated tissues. For

example, the health of the tissue or neuron under study is more easily maintained and monitored. Furthermore, the

neuron can be examined in its normal ionic and cellular microenvironment, with its normal complement of afferent

connections intact. In addition, neurons recorded in vivo are more likely to be spontaneously active, facilitating the

use of extracellular recordings and investigations into the actions of inhibitory neurotransmitters.

With respect to psychopharmacological research, the in vivo preparation provides the most direct link between

neurophysiology and behavior. A drug that elicits a characteristic behavioral response can be administered

systemically to examine how the drug affects neurons that are likely to participate in the behavioral response. For

similar reasons, this preparation is also the most effective for investigating the mode of action of psychoactive drugs

on specific neuronal systems. Although the precise locus of action through which the systemically administered drugs

achieve these effects may be difficult to determine directly, whether a given drug ultimately influences the activity of

a neuronal system of interest can be determined.

Experimental parameters present difficulties that, although not insurmountable, add complexity to the experimental

paradigm. For example, the researcher cannot visually identify the nucleus or the cell to be recorded and must often

rely on indirect techniques for cell identification. However, methods are available to enhance the ability to identify cell

types. Thus, unlike the in vitro preparations, cells may be identified with respect to the projection sites of their axons

by employing antidromic activation—that is, stimulation of the axon terminal region to evoke an action potential that

is conducted back down the axon and subsequently recorded at the soma. Furthermore, by using in vivo intracellular

recording, the neuron in question may be stained with dye and its location, morphology, and neurotransmitter content

identified post hoc by various histochemical and immunocytochemical techniques (e.g., Grace and Bunney 1983; Onn

et al. 1994). In addition, although the precise locus of action of systemically administered drugs cannot be

determined, the drug effects obtained can be compared with those produced by directly applying the drug to the

neuron through microiontophoresis (Figure 5–7) (Bloom 1974). With microiontophoresis, the drug is applied locally to

the recording site, affecting the soma and proximal dendrites of the neuron recorded. However, this may be a

shortcoming when the afferents to be examined synapse distally on the dendritic tree of the neuron.

FIGURE 5–7. Determining the effects of systemic and direct drug administration on neuronal activity.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

14 of 19

10/05/2009 15:49

There are several means of applying drugs to a neuron to examine their actions. During in vivo recording, drugs may be

administered systemically (i.e., intravenously, intraperitoneally, subcutaneously, intraventricularly, intramuscularly) or directly to

the neuron by microiontophoresis or pressure ejection. (A) Systemic administration of a drug is useful for determining how a drug

affects neurons in the intact organism, regardless of whether the action is direct or indirect. In this case, intravenous

administration of the -aminobutyric acid (GABA) agonist muscimol (solid arrows) causes a dose-dependent increase in the firing

rate of this dopamine-containing neuron. (B) In contrast, direct administration of a drug to a neuron will provide information

about the site of action of the drug, at least as it concerns the discharge of the neuron under study. In this case, GABA is

administered directly to a dopamine neuron by microiontophoresis. In this technique, several drug-containing pipettes are

attached to the recording electrode. The pH of the drug solutions is adjusted to ensure that the drug molecules are in a charged

state (e.g., GABA is used at pH = 4.0 to give it a positive charge), and the drug is ejected from the pipette tip by applying very

small currents to the drug-containing pipette. Because the total diameter of the microiontophoretic pipette tip is only about 5 m,

the drugs ejected typically affect only the cell being recorded. In this case, GABA is applied to a dopamine neuron by

microiontophoresis; the horizontal bars show the time during which the current is applied to the drug-containing pipette, and the

amplitude of the current (indicated in nA) is listed above each bar. Note that, unlike the excitatory effects produced by a

systemically administered GABA agonist in (A), direct application of GABA will inhibit dopamine neurons. This has been shown to

be caused by inhibition of a much more GABA-sensitive inhibitory interneuron by the systemically administered drug and

illustrates the need to compare systemic drug administration with direct drug administration to ascertain the site of action of the

drug of interest.

Source. Adapted from Grace AA, Bunney BS: “Opposing Effects of Striatonigral Feedback Pathways on Midbrain Dopamine Cell

Activity.” Brain Research 333:271–284, 1985. Copyright 1985, Elsevier. Used with permission.

A technique that overcomes this shortcoming is the use of combined microdialysis and intracellular recording. In this

approach, a microdialysis probe is used for delivering a drug to the region surrounding the neuron (Figure 5–8). To

preserve the tissue surrounding the probe, the probe is lowered at a rate of 3–6 microns per second using a

micromanipulator (West et al. 2002b). The microdialysis probe is then allowed to equilibrate for approximately 2–3

hours, and sharp-electrode intracellular recordings are conducted within 500 micrometers of the active surface of the

probe. Provided that the probe has been inserted with care, the passive membrane properties, spontaneous spike

activity, and spike characteristics of striatal and cortical neurons recorded during perfusion of artificial cerebrospinal

fluid are found to be similar to those of neurons recorded in animals without microdialysis. The viability of neurons

recorded proximal to the microdialysis probe is further evidenced by the increase in membrane excitability and

spontaneous activity occurring within minutes after introduction of excitatory amino acid agonists (glutamate,

N-methyl-D-aspartate [NMDA]) or the GABAA receptor antagonist bicuculline into the perfusate. Conversely, local

reverse dialysis of tetrodotoxin eliminates action potentials and the spontaneous plateau depolarizations in prefrontal

cortex neurons, indicating that these properties are dependent on synaptic inputs to these neurons. Given these

findings, it is clear that this combination method has unique properties in comparison with local application via

microiontophoresis. Thus, the site of application will span several hundred micrometers and thereby affect a

microcircuit, including the distal dendrites and neighboring neurons of the neuron impaled. Moreover, by applying a

neurotransmitter antagonist, one can ascertain the baseline effect of spontaneous neurotransmitter action on a

neuron being recorded (West and Grace 2002). This approach has recently been applied to single-unit extracellular

recordings (West et al. 2002a) and field potential recordings (Lavin et al. 2005; Goto and Grace 2005a, 2005b).

FIGURE 5–8. Use of a microdialysis probe for delivering drugs locally during in vivo recordings to affect local circuits.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

15 of 19

10/05/2009 15:49

(A) In this schematic diagram, the relationship between the microdialysis probe and the intracellular recording electrode is

depicted. In this case, the neuron recorded is in the striatum. The active surface of the microdialysis probe is shown in gray; this

is the area through which the compound is delivered. The probe is implanted very slowly so as not to disrupt the tissue (i.e., 3–6

m per second) and is perfused with artificial cerebrospinal fluid for 2–4 hours to allow equilibration and settling of the tissue prior

to recording. The intracellular recording electrode is then advanced, and a neuron is impaled. After recording baseline activity for

10 minutes, the perfusate is changed to a drug-containing solution to examine the effects on the neuron. (B) The histology taken

after the recording shows the track of the microdialysis probe; the termination site of the probe tip is indicated by a dashed arrow.

To confirm that the neuron recorded was near the probe, the neuron is filled with a stain (in this case, biocytin) so as to allow

visualization of the neuron. In this case, the neuron was confirmed to be a medium spiny striatal neuron (magnified in insert). ac

= anterior commissure. (C) Recordings taken from the neuron labeled in B. The top trace shows the activity of the neuron while

the microdialysis probe is being perfused with artificial cerebrospinal fluid. The neuron demonstrates a healthy resting membrane

potential, and spontaneously occurring postsynaptic potentials are evident. The lower trace shows the same neuron 15 minutes

after switching to a perfusate containing the dopamine D2 antagonist eticlopride. The neuron shows a strong depolarization of the

resting potential (by 12 mV) as well as increased postsynaptic potential activity and spontaneous spike firing. Since the eticlopride

is blocking the effects of dopamine that is being released spontaneously from dopamine terminals in this region, we can conclude

that basal levels of dopamine D2 receptor stimulation cause a tonic hyperpolarization of the neuronal membrane and suppress

spontaneous excitatory postsynaptic potentials.

Source. Adapted from West AR, Grace AA “Opposite Influences of Endogenous Dopamine D1 and D2 Receptor Activation on Activity

States and Electrophysiological Properties of Striatal Neurons: Studies Combining In Vivo Intracellular Recordings and Reverse

Microdialysis.” Journal of Neuroscience 22:294–304, 2002. Copyright 2002, Society for Neuroscience. Used with permission.

On the other hand, the properties that confer distinct advantages on the in vivo preparation with respect to examining

how drugs act in the intact organism also limit the type of data that may be collected. Regarding drug administration,

some drugs are not easily applied via microdialysis or do not readily cross the blood–brain barrier, or they may

actually produce their direct actions outside of the brain via an effect on peripheral organs. Thus, although dopamine

cells can be excited by microiontophoretic administration of cholecystokinin (Skirboll et al. 1981), the excitation

produced by systemic administration of this peptide is mediated peripherally and affects the brain via the vagusPrint: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

16 of 19

10/05/2009 15:49

(Hommer et al. 1985). In addition, the inability to control the microenvironment of the neuron restricts the analysis of

the ionic mechanisms underlying cell firing or drug action because the researcher cannot readily control the precise

drug concentration or the ionic composition of the fluid surrounding the neuron. There is also difficulty in segregating

local actions of drugs versus those imposed on afferent neurons or their local axon terminals. Therefore, whereas the

in vivo preparation affords many advantages with respect to examining how behaviorally or therapeutically effective

drugs may exert their actions through defined neuronal systems, examination of the site of action or the membrane

mechanisms underlying these responses is more readily accessible with in vitro systems.

In Vitro Electrophysiological Recordings From Brain Slices

Recordings of neurons maintained in vitro have led to significant advances in understanding the ionic mechanisms

underlying neurotransmitter and drug actions. This preparation consists of slices 300–400 m thick cut from the brain

of an animal soon after decapitation. If this procedure is done carefully and the brain slices are rapidly placed into

oxygenated physiological saline, the neurons within the slices will remain alive and healthy, often for 10 hours or

more. Because the neurons are recorded in a chamber with oxygenated media superfused over the slice, several

advantages may be realized:

1. Both intracellular and extracellular recordings are more stable because blood and breathing pulsations are absent.
2. Visual control over electrode placement is achieved.
3. The ionic composition of the microenvironment may be controlled precisely.

Little interference from the activity of long-loop afferents occurs, and the near-absence of spontaneous spike discharge limits

the contribution of local circuit neurons to the responses.

4.  

Furthermore, in contrast to microiontophoresis, the concentration of drug in the solution can be controlled precisely.

This preparation is also the most complex that can be used for patch clamp recordings because debris may be removed

and the patch pipette placed on selected neurons under visual control with a high-resolution optics system (Edwards

et al. 1989).

Nonetheless, because of the isolated nature of this system, the results obtained may not precisely reflect the

physiology of the intact system. For example, dopamine neurons recorded in vivo have been characterized by their

burst-firing discharge pattern (Grace and Bunney 1984a), which appears to be important in regulating

neurotransmitter release (Gonon 1988). However, dopamine neurons recorded in vitro do not fire in bursting patterns

(Grace and Onn 1989) (Figure 5–9). On the other hand, this distinction provides what may be an ideal system for

examining the factors that cause in vivo burst firing. Therefore, the most complete model of the functioning of a

system or of its response to drug application can be derived by comparing the results obtained in vitro with those in

the intact organism.

FIGURE 5–9. Variation (sometimes substantial) in patterns of activity of a neuron type, depending on the preparation

in which it is recorded.

(A) Extracellular recordings of a dopamine neuron in an intact anesthetized rat (i.e., in vivo) illustrate the typical irregular firing

pattern of the cell, with single spikes occurring intermixed with bursts of action potentials. (B) In contrast, intracellular recordings

of a dopamine neuron in an isolated brain slice preparation (i.e., in vitro) illustrate the pacemaker pattern that occurs exclusively

in identified dopamine neurons in this preparation. For dopamine neurons, a pacemaker firing pattern is rarely observed in vivo,

and burst firing has never been observed in the in vitro preparation. However, although the activity recorded in vitro is obviously

an abstraction compared with the firing pattern of this neuron in vivo, a comparative study in each preparation does provide the

opportunity to examine factors that may underlie the modulation of firing pattern in this neuronal type.

Source. Adapted from Grace AA: “The Regulation of Dopamine Neuron Activity as Determined by In Vivo and In Vitro Intracellular

Recordings,” in The Neurophysiology of Dopamine Systems. Edited by Chiodo LA, Freeman AS. Detroit, MI, Lake Shore

Publications, 1987, pp. 1–67 (Copyright 1987, Lake Shore Publications. Used with permission); and Grace AA, Bunney BS:

“Intracellular and Extracellular Electrophysiology of Nigral Dopaminergic Neurons, I: Identification and Characterization.”

Neuroscience 10:301–315, 1983. Copyright 1983, International Brain Research Organization. Used with permission.Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

17 of 19

10/05/2009 15:49

Recordings From Dissociated Neurons and Neuronal Cell Cultures

Recordings from isolated neurons are actually a subset of in vitro recording methods, with many of the same

advantages in terms of accessibility and stability. Furthermore, because the neurons can be completely visualized,

advanced techniques such as patch clamping are more easily done. A unique advantage of this system can be obtained

by coculturing different neuronal populations. For example, defining the effects of a noradrenergic synapse on a

hippocampal pyramidal neuron more precisely may be possible by coculturing these cell types and allowing them to

make synapses. In this way, the researcher has visual control over impaling neurons that constitute a presynaptic and

postsynaptic pairing. On the other hand, the synapses formed are not necessarily limited to those that occur naturally

in the intact organism, in terms of both the location of the synapse on the neuron and the classes of neurons that are

interconnected. Furthermore, the altered neuronal morphology present in these preparations may modify the response

of the neurons to drugs. Nonetheless, when the analysis is limited to well-defined responses, such as

second-messenger actions or ion channel measures, this system affords an unparalleled level of accessibility.

RELATIONSHIP BETWEEN BIOCHEMICAL AND ELECTROPHYSIOLOGICAL MEASURES OF

NEURONAL ACTIVITY

The methods outlined here are directed at analyzing the activity of individual neurons as a means of assessing their

role in pharmacological responses or behavioral actions. This is based on the premise that the discharge of a neuron in

some manner reflects its release of a transmitter onto a postsynaptically located target neuron. As such, biochemical

measures of neurotransmitter levels would be predicted to correspond to the activity changes occurring during

electrophysiological recordings from neurons (Roth 1987). In several cases, such approaches have helped to define

the physiological relevance of recorded neuron activity. One case in which this has proven valuable is in the analysis of

firing pattern. For example, dopamine neurons, like many other cell types in the central nervous system, are capable

of discharging trains of action potentials in two patterns of activity: single spiking and burst firing. However, their

range of firing rates is comparatively restricted, with most cells firing only between 2 and 8 Hz. Nonetheless,

information on the temporal relationship between spikes and bursts (Grace and Bunney 1984a) has been used in in

vivo voltammetry studies to measure dopamine levels. Dopamine cells firing in bursts will release two to three times

more neurotransmitter per spike from their terminals than those discharging at similar frequencies but in a steady

firing pattern (Gonon 1988). Therefore, in this case, knowledge of the physiological firing pattern provided

information to the electrochemist that resulted in the elucidation of the physiological consequence of burst firing in

this system.

However, the extrapolation between biochemical and electrophysiological measures may not always be valid. Thus,

recordings from single neurons may not necessarily reflect the activity across the population of neurons of interest.

Therefore, a drug that exerts an action via activation of the nonfiring population of neurons may be overlooked if its

actions are assessed on single spontaneously discharging neurons (Bunney and Grace 1978; Grace and Bunney

1984b). Furthermore, the response may be confined to a topographically defined subset of neurons mediating a

particular response (e.g., a change in the activity of neurons regulating movement of the leg would not be predicted if

the response involves a reaching movement of the arm). With respect to biochemical measurements, actions of

transmitters at presynaptic terminals could dramatically alter the amount of neurotransmitter they release

independent of neuronal discharge (e.g., Grace 1991). On the other hand, electrophysiological measurements enable

researchers to examine responses that occur very rapidly. Indeed, a massive but transient activation of spike

discharge in a neuronal system may evoke a substantial behavioral response, whereas biochemical measurements of

neurotransmitter release performed over a long time course may dilute the impact of the transient event. Therefore,

although the results obtained from each measure may not be directly comparable, the electrophysiological

measurements are better optimized for detecting transient events.

SUMMARY

In this chapter, we reviewed several electrophysiological techniques and preparations used in the analysis of nervous

system function. Each approach is characterized by a set of unique advantages and potential shortcomings inherent in

the method. Nonetheless, it should be apparent that no single technique has an overwhelming advantage in

psychopharmacological research. Instead, by matching the preparation to the question at hand, and through the

judicious comparison of data obtained from intact versus isolated preparations, the various limitations may be

systematically overcome to yield a more broadly applicable model of psychopharmacological action.

REFERENCES

Bloom FE: To spritz or not to spritz: the doubtful value of aimless iontophoresis. Life Sci 14:1819–1834, 1974

[PubMed]

Bunney BS, Grace AA: Acute and chronic haloperidol treatment: comparison of effects on nigral dopaminergic cell

activity. Life Sci 23:1715–1728, 1978 [PubMed]

Bunney BS, Walters JR, Roth RH, et al: Dopaminergic neurons: effect of antipsychotic drugs and amphetamine on

single cell activity. J Pharmacol Exp Ther 185:560–571, 1973 [PubMed]

Edwards FA, Konnerth A, Sakmann B, et al: A thin slice preparation for patch clamp recordings from neurons of the

mammalian central nervous system. Pflugers Arch 414:600–612, 1989 [PubMed]Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

18 of 19

10/05/2009 15:49

Freedman R, Hoffer BJ, Woodward DJ, et al: Interaction of norepinephrine with cerebellar activity evoked by mossy

and climbing fibers. Exp Neurol 55:269–288, 1977 [PubMed]

Gonon FG: Nonlinear relationship between impulse flow and dopamine released by rat midbrain dopaminergic neurons

as studied by in vivo electrochemistry. Neuroscience 24:19–28, 1988 [PubMed]

Goto Y, Grace AA: Dopaminergic modulation of limbic and cortical drive of nucleus accumbens in goal-directed

behavior. Nat Neurosci 8:805–812, 2005a

Goto Y, Grace AA: Dopamine-dependent interactions between limbic and prefrontal cortical plasticity in the nucleus

accumbens: disruption by cocaine sensitization. Neuron 47:255–266, 2005b

Grace AA: Phasic versus tonic dopamine release and the modulation of dopamine system responsivity: a hypothesis for

the etiology of schizophrenia. Neuroscience 41:1–24, 1991 [PubMed]

Grace AA: The depolarization block hypothesis of neuroleptic action: implications for the etiology and treatment of

schizophrenia. J Neural Transm 36 (suppl):91–131, 1992

Grace AA, Bunney BS: Intracellular and extracellular electrophysiology of nigral dopaminergic neurons, I:

identification and characterization. Neuroscience 10:301–315, 1983 [PubMed]

Grace AA, Bunney BS: The control of firing pattern in nigral dopamine neurons: burst firing. J Neurosci 4:2877–2890,

1984a

Grace AA, Bunney BS: The control of firing pattern in nigral dopamine neurons: single spike firing. J Neurosci

4:2866–2876, 1984b

Grace AA, Bunney BS: Opposing effects of striatonigral feedback pathways on midbrain dopamine cell activity. Brain

Res 333:271–284, 1985 [PubMed]

Grace AA, Llinás R: Dehydration-induced morphological artifacts in intracellularly stained neurons: circumvention

using rapid DMSO clearing. Neuroscience 16:461–475, 1985 [PubMed]

Grace AA, Onn SP: Morphology and electrophysiological properties of immunocytochemically identified rat dopamine

neurons recorded in vitro. J Neurosci 9:3463–3481, 1989 [PubMed]

Hommer DW, Palkovits M, Crawley JN, et al: Cholecystokinin-induced excitation in the substantia nigra: evidence for

peripheral and central components. J Neurosci 5:1387–1392, 1985 [PubMed]

Kandel ER: A Cell-Biological Approach to Learning (Grass Lecture Monograph 1). Bethesda, MD, Society for

Neuroscience, 1978, pp 1–90

Lavin A, Nogueira L, Lapish CC, et al: Mesocortical dopamine neurons operate in distinct temporal domains using

multimodal signals. J Neurosci 25:5013–5023, 2005 [PubMed]

Llinás RR: The intrinsic electrophysiological properties of mammalian neurons: a new insight into CNS function.

Science 242:1654–1664, 1988 [PubMed]

Onn SP, Berger TW, Grace AA: Identification and characterization of striatal cell subtypes using in vivo intracellular

recording and dye-labelling in rats, III: morphological correlates and compartmental localization. Synapse

16:231–254, 1994 [PubMed]

Roth RH: Biochemical correlates of the electrophysiological activity of dopaminergic neurons: reflections on two

decades of collaboration with electrophysiologists, in Neurophysiology of Dopaminergic Systems—Current Status and

Clinical Perspectives. Edited by Chiodo LA, Freeman AS. Detroit, MI, Lake Shore Publications, 1987, pp 187–203

Skirboll LR, Grace AA, Hommer DW, et al: Peptide-monoamine coexistence: studies of the actions of a

cholecystokinin-like peptide on the electrical activity of midbrain dopamine neurons. Neuroscience 6:2111–2124, 1981

[PubMed]

Terzuolo CA, Araki T: An analysis of intra- versus extracellular potential changes associated with activity of single

spinal motoneurons. Ann N Y Acad Sci 94:547–558, 1961 [PubMed]

West AR, Grace AA: Opposite influences of endogenous dopamine D1 and D2 receptor activation on activity states and

electrophysiological properties of striatal neurons: studies combining in vivo intracellular recordings and reverse

microdialysis. J Neurosci 22:294–304, 2002 [PubMed]

West AR, Galloway, MP, Grace AA: Regulation of striatal dopamine neurotransmission by nitric oxide: effector

pathways and signaling mechanisms. Synapse 44:227–245, 2002a

West AR, Moore H, Grace AA: Direct examination of local regulation of membrane activity in striatal and prefrontal

cortical neurons in vivo using simultaneous intracellular recording and microdialysis. J Pharmacol Exp Ther

301:867–877, 2002b

Woodward DJ, Moises HC, Waterhouse BD, et al: Modulatory actions of norepinephrine in the central nervous system.

Fed Proc 38:2109–2116, 1979 [PubMed]Print: Chapter 5. Electrophysiology

http://www.psychiatryonline.com/popup.aspx?aID=408408&print=yes…

19 of 19

10/05/2009 15:49

Copyright © 2009 American Psychiatric Publishing, Inc. All Rights Reserved.