Detection of dopamine dynamics in the brain - Analytical Chemistry

---

Detection of DoDamine Dynamic;

\ -

&I

Dendrite

I?.Mark Wightman, Leslie J. May, and Adrian C. Michael Department of Chemistry Indiana University Bloomington. IN 47405 Twelve years ago, Ralph Adams puhlished an article in ANALYTICAL CHEMISTRYconcerning brain chemistry ( I ) . He described a number of electroanalytical techniques that were heing developed to investigate the role of neurotransmitters in the brain. That article inspired many people to get involved in this exciting new area of analytical chemistry. In this REPORT we will focus on the use of in vivo probes, especially voltammetric microelectrodes, to monitor dynamics of neurochemical events in the extracellular space of the brain. Adams’s early work raised several issues regarding the interpretation of such measurements, and many of these have now been resolved. At the same time, our increased understanding of the challenges of research in the brain has led to the definition of a series of new questions that require improved analytical methods. In this REPORT we will show some of the ways in which these questions have been approached. 0003-2700/88/0360-769A/$O1.50/0 @ 1988 American Chemical Society

However, it is not our intention to provide a complete review; so much has been done by so many laboratories that the reader is referred to more extensive reviews for a general overview. Neurotransmission Understanding more ahont the structure and function of the brain is one of the biggest challenges for modern scientists. This organ is being studied hy a broad range of investigators collectively called neuroscientists. How can an analytical chemist aid in nnderstanding the brain? The answer that Ralph

Adams proposed was the development of methods to analyze the small molecules found in the brain known as neurotransmitters. T o design such techniques, one has to have an understanding of the environment in which these compounds perform their regulatory actions. In this way we can approach the goal of designing a sensor that provides dynamic information concerning chemical concentration changes. The most interesting cells in the

brain are neurons that have developed the specialty of collecting, integrating, and relaying information. A diagram of a neuron is shown in Figure 1. In reality, neurons come in a variety of shapes and sizes, hut the “typical” neuron shown contains the elements needed to understand their basic function. The cell body has two types of projections: the dendrites, which serve as an input for information, and the axon, which serves as an output device. Information is conveyed in the form of electrical potentials across the cell membrane. Normally, the inside of the neuron is negative with respect to the outside because of an unequal distribution of ions between the two regions. The opening of specialized ion channels causes the membrane to depolarize. If the depolarization is sufficiently great, this results in an action potential-a rapid voltage change that propagates down the axon to the terminals. The propagation rate can exceed 50 mfs in some neurons, resulting in the rapid propagation of information from one neuron to another. The role of neurotransmitters is to provide the communicating link hetween neurons. Electron micrographs of the terminal region of neurons, along with other evidence, have led to the

ANALYTICAL CHEMISTRY. VOL. 60, NO. 13. JULY 1, 1988

769A

Flgure 2. The terminal (or synaptic) region coupling two cells. (SYN = synU’msis: DA = dopamine: V = dopamine storage vesicles: N = nwonal release: SC = synap tic cleft R = pe and poDtsynaptic dopamine receptors: U = dopamine uptake: SO = synaptic overtbw: M = metabolic enzymes.)

view shown schematically in Figure 2. Neurotransmitters are stored in vesicles in the nerve terminal region at a concentration that may be as high as 1M. The arrival of an action potential a t the terminal leads to the secretion of the neurotransmitter into the synaptic cleft. The neurotransmitter diffuses a c r w the synaptic cleft and interacts with receptors on the membrane of the next neuron down the information chain. Receptors are proteins that recognize the presence of particular neurotransmitters; in response to their presence, ion channels are opened or closed, thus affecting the potential of

the target neuron. Electrophysiologists have developed microelectrodes to monitor these potential changes and have found that these events occur on a millisecond time scale. A variety of mechanisms exist to remove the neurotransmitter from the synaptic region so that thenext action potentialcan initiate this series of events again. To better understand the process of neurotransmission and to learn about the processes that modulate these events. we would ideally like to monitor chemical dynamics in the synaptic cleft with millisecond time resolution. Spatial and temporal requirements make *1 /I/

this a virtually impossible task, but considerable information can be obtained with methods already developed. In this article we will focus on the specific neurotransmitter dopamine, which is a target molecule used by several laboratories. The observation that dopamine is easily oxidized, rendering it suitable for electroanalysis, prompted Adams to initiate work in this field. Dopamine was discovered to be a neurotransmitter in the late 1950s. Using a fluorescence assay, neuroscientists were able to show that dopamine is found in high amounts (50 nmol/g) in a region of the brain known as the caudate nucleus. In the 1960s it was found that patients with Parkinson’s disease show an almost complete depletion of dopamine in this region. The pathways of dopamine-containing neurons have been mapped in the rat brain. One of the major pathways is shown in Figure 3. Later studies showed that neurons in the caudate nucleus can accumulate dopamine from the surrounding fluid, a process referred to as uptake; this process is likely to be the mechanism of inactivation of dopamine neurotransmission. It was further shown that drugs of abuse such as cocaine and amphetamines could inhibit this uptake process; this cellular action correlates with the behavioral changes induced by these drugs. A correlation also exists between the binding of antipsychotic drugs to dopamine receptors and their efficacy in controlling the symptoms of schizophrenia. The interest in dopamine neurotransmission is far reaching, and it seems that the proper regulation of this process is required for normal functioning and behavior. The ability to monitor dopamine neurotransmission directly could provide useful insights into this process, and this fundamental information may in turn be beneficial in the design of therapies for diseases associated with dopamine systems. For example, microvoltammetric electrodes can be used to monitor the capability of cell transplants to release catecholamines (2). This type of preparation is currently being evaluated as a “brain transplant” for Parkinson’s disease patients. Because this disease arises from an abnormal loss of dopamine-containing neurons, it has been proposed that a transplant of catecholamine-secreting cells into the damaged region could alleviate the symptoms. The capability of the cells to release catecholamines before and after transplant can provide information necessary for evaluating the efficacy of this approach. Analytical challenges The design of chemical sensors capable of measuring neurotransmitters a t the site of their action requires consider-

7701

*

ANALYTICAL CHEMISTRY. VOL. 60, NO. 13, JULY 1. 1988

1

It works like adream. Exchanging one LC column for another instantly and effortlessly It's a Rheodyne column selection valve-and it eliminatesthe need to physically discmnect a column and reconn nectanotherinits place. You simply turnthe valve

-

handle to redirect flow from one column to another. Our Model 7ooovalve selects one of two columns (see photo and left diagram). Two of our Model 7060 valves select any of five columns or a bypass flushing ClRCCE 140 ON READER SERVICE CARD

line (see right

u.

Ebr technical notes and product literature, phone your Rheodyne

dealer. Or if more convenient, address Rheodyne, Inc.,EO. Box 996,Cotati, California 94928, USA. Phone (707)664-905r

RWODYNG TrlE -C COhhECT ON COMPANY

ation of all of the factors described above as well as many others. Many researchers have chosen the caudate nucleus as a site to test voltammetric sensors. The rat has been used for experimental purposes because a great deal is known about the physiology, pharmacology, and anatomy of the rat caudate nucleus as a result of its widespread use as a laboratory animal. For example, it is known that -20% of the volume of the rat brain is occupied by the extracellular fluid (3). This aqueous fluid has an ionic strength comparable to seawater and thus is a suitable medium for voltammetry. These factors, coupled with the relatively large sizeofthe caudatenucleusand the high amounts of dopamine present, led to the initial investigations with in vivo electrodes in that region. The early experiments from Adams’s lab showed that distinct voltammetric waves could he obtained from electrodes implanted in the caudate nucleus. Carbon electrodes were used because carbon tends to he more compatible with biological tissues than other commonly used electrode materials. Many questions were raised by the preliminary experiments: What are the compounds giving rise to the voltammetric waves? Are neurotransmitters detected? Can the measured signal he related to concentrations in the extracellular fluid? * What temporal resolution is required for meaningful measurements? What is the usable lifetime of an electrode in the brain? What is the diffusional environment at the electrode tip? Where is the electrode with respect to the synapse? * What does the observation of a neurotransmitter mean with respect to neurotransmission? The first four questions are the type that analytical chemists often ask when making new measurements. The last two, however, are unique to the heterogeneous environment that exists inside the brain. In order to get as close to the synaptic region as possible, and to minimize damage to the brain tissue, a variety of microelectrodes have been designed. Electrodes fabricated from carbon fibers (10-fim diameter) have been widely used (4-6); an example of such an electrode is shown in Figure 4. These electrodes are not small enough to be inserted inside the synapse, hut they are sufficiently small that the measurements reflect the chemical composition of the extracellular fluid surrounding the neurons. Identifying electroactive species in the rat caudate nucleus Voltammograms for dopamine (7)in a buffer comparable to the composition of extracellular fluid (pH 7.4) show a

Figure 4. Electron micrograph of the tip of a Nafion-coated. carbon fiber electrode.

quasi-reversihle, two-electron oxidation wave with an E1,z of 0.12 V a t a carbon electrode versus a saturated calomel electrode (WE). A similar response can he obtained for dopamine injected into the brain adjacent to an implanted voltammetric electrode (8). Early experiments demonstrated that carbon electrodes could provide voltammetric information in vivo and that the electrode surface was not destroyed by implantation. However, voltammograms obtained in the caudate nucleus of an anesthetized rat show irreversible waves. The wave observed in vivo that might contain dopamine (peak potential of 4 . 4 V) dramatically decreases in amplitude when the enzyme ascorbic acid oxidase is injected near the voltammetric electrode (9), showing that ascorbic acid is a major component of the voltammogram. A dramatic increase in the concentration of electroactive species associated with this wave occurs following administration of drugs such as amphetamines, which are thought to increase the concentration of dopamine in extracellular fluid (10). The broad voltammetric wave of ascorbate overlaps severely with that of dopamine, complicating the interpretation of this result. Voltammograms of ascorbate a t carbon electrodes exhibit drawn-out waves characteristic of slow, two-electron transfer with an Ellzof 4 . 2 V (7). In contrast, much faster electron transfer occurs at metal electrodes and an E1/2 of --0.08 V is obtained. In 1980 Gonon et al. showed that large-amplitude electrochemical pretreatment (+3.0 V) of a carbon fiber electrode al-

7 7 2 A * ANALYTICAL CHEMISTRY, VOL. 60, NO. 13, JULY 1, 1988

ters the voltammetric behavior of ascorbate (11)so that it is oxidized a t a potential similar to that observed a t metal electrodes (Figure 5). The origin of these surface changes has been examined in some detail (12). Following pretreatment, dopamine and ascorbate can be voltammetrically resolved a t carbon electrodes. With improved voltammetric resolution it was shown that ascorbate was the substance that increased in concentration in the brain in response to amphetamine (11, 13). A distinct wave is observed a t the potential where dopamine is oxidized, hut this is attributed to dihydroxyphenylacetic acid (DOPAC, a metabolite of dopamine) on the hasis of the effect of pharmacological agents on the wave amplitude. These results were quite startling to neuroscientists and electrochemists alike. The alteration of ascorbate concentration in response to a dopaminergic drug was unexpected and suggested a previously unrecognized role for ascorbate in neurotransmission, an area that has subsequently been investigated in moredetail (14-16). Furthermore, it raised the question of whether other easily oxidized substances might be contributing to the voltammetrically measured signal. Thus an inventory of the compounds in the extracellular fluid was required. Exhacellular fluid sampling Before the development of in vivo electrochemistry, perfusion methods had been developed to sample the extracellular space of the brain (17).The original “push-pull” cannula consists of

First we made a Solid Phase Extraction tube for highly reproducible extractions When you buy one of our Supelclean" Solid Phase Extraction tubes, you get more than just a tube. You get real reproducibility Because we manufacture our Solid Phase Extraction tubes by monomerically bonding the silica-basedpacking to minimize variability in the surface coverage. And before we pack our tubes, we test every lot to make certain each one conforms our quality specifications.

Next we backed it with expert technical service After you buy the tube, you get the benefit of our technical service department This busy group of experts has solved hundreds of difficult problems for chromatographers all over the world. As a Supelco customer, you can call them and get advice on virtdally any solid phase extraction problem in only a few minutes - just as though you had your own personal chromatography consultant.

Then we wrote the book on solid phase extraction

9

SUPELCo

A R W M AND UAAE COMPANY

Supalco Park Bellefonte, PA 16823-0048 Phone [814]359-3441and [814]359-3446 TWX 510 670-3600 FAX 814-359-3044

BLIPELGO, INC.

Now you can even have our Guide to Solid Phase Extraction. Those in the know have told us this is the best solid phase extraction guide they've seen. We want you to have the opportunity to judge for yourself - absolutely free - and at the same time find out more about our reproducible Supelclean solid phase extraction tubes. To get your tree copy of our Guide to W/dPhase Extraction,slm@ycalt OurtechM 88tvic6 *PIemt 359-3441) before August 31, and m e n t h tMs ad. It wouldn't hurt to cell right r#mr whHe supplies last.

CIRCLE 152 ON READER SERVICE CARD

-

I

+0.6

I

I

+0.2

I

I I -0.2

I

I

I

+0.6

I

I

+0.2

I

- 0.2

E(VW.SCE)

iura 5. Voltammogramsof dopamine (DA) (left) and ascwbic acid (AA) (right)ai cylindrical carbon fiber electrode in pH 7.4 solutions. In each case voftnmogam 1 was r d at m tmmated elects& and voltammogram 2 was reoarded following e W m i c a l prebeabnenl. Scan rate: 100 mV 9-l; concenlretlon 01 DA 0.1 mM, wncenblllbn 01 AA: 1.0 mM. (Adapted vim penlarim hom Relerence 7.)

two concentric stainless steel needles through which physiological fluid is simultaneously infused (pushed) and withdrawn (pulled). In this way the compounds in the brain are removed and can be analyzed using any technique suitable for use with small volumes. In early studies, radiolabeled substances were used because of this volume restriction. In later studies liquid chromatography with electrochemical detection (LC-EC) was used; its high sensitivity makes it ideal for these analyses. Using small-bore columns, ferntogram quantities in nanoliter volumes may be determined (18). The chromatographic separation also provides the chemical resolution lacking in in vivo voltammetric methods. The information obtained from this experiment provided the inventory of electroactive substances that is necessary to interpret in vivo voltammetric results. With this method, Justice and eo-workers confirmed the results of Gonon concerning ascorbate (19). It was shown that uric acid, homovanillic acid, and DOPAC were other electroactive substances present in the extracellular fluid of the caudate nucleus, and the concentrations observed were similar to those estimated by voltammetric measurements. In these studies dopamine was undetectable. An improvement in the push-pull method was first devised by Ungerstedt (In,who designed a dialysis perfusion cannula This device is similar to the original cannula except that the flow of perfusion fluid is enclosed within a dialysis membrane. Small molecules present in the external medium can diffuse across the membrane and 774A

*

be removed from the brain. Several designs of dialysis cannulae have since been introduced. A miniaturized design that incorporates small-diameter fused-silica tubes inserted into a dialysis hollow fiber of 300-fim 0.d. is shown in Figure 6. The small internal volume of this cannula allows the use of extremely slow perfusion flow rates (-100 nL/min). Near-equilibrium recovery is obtained at slow perfusion flow rates, which means that the rate of m w transfer through the external medium has little effect on the amount of material present in the dialysate. This allows calibration of tissue concentrations by comparison with aqueous standard solutions, despite the very different mass transport characteristics of these media. With slow perfusion rates the amount of material removed from the external medium per unit of time is reduced, suggesting that the chemical perturbation of the living tissue caused by the sampling is also reduced (20,21).In addition, histological studies show that the damage to the brain tissue is considerably less when the perfusate is enclosed in the dialysis tubing. The dialysis membrane also provides a degree of sample cleanup because of the molecular-weight cutoff of the membranes employed (typically 30005000 amu). The elimination of manual sample cleanup procedures has led to the development of integrated and automated perfusion-analysis systems (22) in which the dialysate is perfused directly into the injection valve of a liquid chromatograph. Thus very small (submicroliter) samples can be used, and the sampling frequency is limited only by the time of the chromatograph-

ANALYTICAL CHEMISTRY, VOL. 60. NO. 13. JULY 1, 1988

Flgure 6. Schematic of dialysis cannula

construction. A 0.9mm 0.d. dlalys1s hollow I l k swroulds two

hsedsllica lubes. portuslon fluid is introduced t h o q h one of me tubes and r e w e d via the om w. small mo11wiu188 In me edema1 soluilm are able Io diffuse a w m s he dlaiysla membrane ( m p l e d with penlarlon tram Reference 20.)

ic separation with small-bore columns (Figure 7). By using this approach, one can detect the low basal concentration (-30 nM) of dopamine in the extracellular fluid of the caudate nucleus. The change in the concentration of dopamine following a pharmacological stimulus known to exert profound behavioral changes can also be detected with dialysis methods (Figure 8). Thus dialysis perfusion coupled to small-bore LC-EC provides a method capable of making quantitative determinations of extracellular fluid dopamine levels without the interference of other electroactive substances on a minute time scale. The high sensitivity and selectivity of the dialysis technique have established it as an important tool for the measurement of changes in brain chemistry. This technique will be more widely used because with it one can sample any compound that crosses the membrane and for which suitable analytical techniques exist. For example, it could be used to measure the rate of penetration of pyschoactive drugs from the bloodstream into the brain. There-

fore the development of other in vivo sensors should be aimed a t providing complementary measurements. The dimensions of dialysis devices are typically large, on the order of millimeters. This makes them applicable to large brain regions such as the caudate, but smaller regions cannot easily be stud. ied. Smaller sensors will provide cam. plementary information on more local concentration changes. Dialysis probes sample on a minute time scale. As seer in Figures 7 and 8, this is adequate foi characterizing pharmacological effeck that may occnr on a time scale of several minutes to hours. Other forms of in vivo sensors should be capable of making faster measurements to provide complementary information related to more rapid methods of neural stimulation. Electrodes with improved selectivity, d i v i i , and temporal respome

The results of the perfusion experiments clearly define the selectivity and sensitivity requirements for an electrode suitable for in vivo detection of dopamine in the caudate nucleus. Discrimination against ascorbate and DOPAC, present at lo4 to 103 times greater concentration, must be provided. The other electroactive substances mentioned are oxidized a t more positive potentials and thus can be distinguished hy voltammetry. The extremely low concentration of dopamine in the extracellular fluid of the caudate nucleus provides a large challenge for detection. Its low concentration is surprising when compared with the levels found by analysis of the tissue. If the caudate nucleus were considered as a homogeneous phase, the concentration would be expected to be 50 pM.A concentration difference of IO3 exists between the inside and outside of dopamine neurons. The use of perflnorinated ion-exchange membranes has led to improved electrodes. Although several groups have described the properties of Nafion (a cation-exchange material) coated on electrodes, Adams's group was the first to show that it can be used to improve selectivity for dopamine (23, 24). The amino side-chain of dopamine is protonated at physiological pH and thus can be accumulated at a coated electrode. In contrast, ascorbate and DOPAC (both weak acids) are anions at this pH and tend to be excluded, resulting in an approximate ZOO-fold greater sensitivity for dopamine as compared with these substances. Furthermore, if the coating is sufficiently thin, dopamine can penetrate through the film on a subsecond time scale (25). Untii the advent of very small electrodes, the preferred methods for rapid electrochemical measurements were potential step techniques such as

0.5

04

.9., ? 03 $ 2 b O2

E 01

I

I

0 10

15

20

25

30

Time (min) Figure 7. Consecutive chromatograms of striatal dialysate obtained at 5-inin lntervals from an unanesthetized rat. The peaks marked with an asterisk are the dopamine peaks for each injected sample. (Adapted with permission from Reference 22.)

chronoamperometry. The successive repetition of short (100 ms), repetitive (two seconds apart) potential steps can be used to provide an almost continuous monitor of the concentration of electroactive species at one potential, and this technique has been employed extensivelv. However, microvoltammetric electrodes facilitate cyclic voltammetry at fast scan rates, and this technique was used by Millar for in vivo recording (26).The concentration of an electroactivesubstance is related to the wave height orarea. I'nfortunately, the

voltammograms contain a considerable contribution from charging current. Because this Contribution can be removed by digital subtraction, this method is best suited to the determination of rapid concentration changes in which the subtraction procedure gives the voltammomam of the sDecies that has changed inconcentrati&. The combined use of fast-scan cyclic voltammetry with Nafion-coated, carbon fiber electrodes provides a method that has many of the characteristics required for an in vivo sensor

Time (min) Figure 6. Average dopamine concentration ObSeNed in the extracellular fluid of unanesthetized rats using dialysislLC-EC (Circles = control: triangles = administration of 10 mglkg cocaine at t = 30 min.) R W S U ~are ~ S from three animals in each CBSB: m w s are representee as standard wms 01 the mean. (Adapted with pmissiOn horn Reference 22.)

ANALYTICAL CHEMISTRY. VOL. 60. NO. 13, JULY 1. 1988 * 775,.

for dopamine. Voltammograms recorded at a scan rate of 300 V s-I take less than 10 ms to acquire. The voltammograms are repeated at 100-msintervals, which is the time scale for dopamine penetration through the film. These electrodes give an almost instantaneous Concentration readout. At fast scan rates the voltammograms provide a unique “fingerprint” for identification because the shapes of the voltammograms are kinetically controlled, and those for DOPAC and ascorbate are quite different from dopamine (27). Because of this, the selectivity for dopamine over ascorbate is now increased to approximately lOM)-fold. The coated electrodes show less deterioration than hare electrodes when implanted in the brain, and continuous use for up to eight hours is commonplace. Because of the low value of the diffusion coefficient in Nafion (D = 1X 10-9 cm2 s-9, the diffusion layer (the region perturbed by electrolysis) is restricted to the interior of the film. This has two advantages: The dopamine-o-quinone produced by the electrolysis of dopamine, a potential neurotoxin, is generated for only a brief period of time and is restricted to the interior of the film and the diffusion-layer region is thi same in the brain and in a beaker whei the electrode is calibrated after the ex periment.

millimolar range and the substance detected was not dopamine hut homovanillic acid (HVA). HVA is a metabolite of dopamine, and the metabolism must have occurred as the released suhstance was transported to the ventricle. With the electrode implanted in tissue, an entirely different view is obtained (Figure 9a). When the stimulus is initiated, a chemical change is detected within 100 ms. The observed concentration, which depends on the stimulus frequency, rapidly increases during the stimulation and then returns to the original baseline within a few seconds after cessation of the stimulation. The rapid time scale of the concentration changes make it clear that suhsecond time resolution is required to observe these events. The detected substance is voltammetrically identical to dopamine (not HVA), as shown in Figure 9h. Much evidence has

Observation ofdopamine in VIVO Despite the improvements in methodology described above, the Nafioncoated electrodes are not yet capable of detecting the low nanomolar concentrations of dopamine that normally exist in the extracellular fluid. Even with extensive use of signal-averaging techniques, the detection limit with fastscan voltammetry is approximately a 100-nM change in dopamine concentration. For this reason, we have implanted stimulating electrodes in the vicinity of the axons of dopamine neurons. These stimulating electrodes can he used to artificially trigger action potentials (28). Voltammetric electrodes in the caudate nucleus are used to si multaneously monitor the evoked re lease of dopamine. This experiment is similar to the one described by Adams ( I ) . However, the results are very different because of the differences in experimental protocol and are attributable to the analytical advances that have occurred in the past 12 years. In the early work, the electrode was not implanted in the tissue: instead, it was in a fluid cavity of the brain, the lateral ventricle, which is adjacent to the caudate nucleus. The spatial separation of the electrode from the release site caused a time delay (-20 min) in the observed chemical response to the stimulation. Furthermore, the detected response was in the 776A

Time(s)

Flgure 9. Data obtained in the caudate nucleus of an anesthetized rat from a Nafioncoated, carbon fiber electrode wlth voltammetry (300 V s-’) during stimulations of dopamine-containing

neurons. (a) Temporal changes obsewed during stlmulations of dlllerent frequmy. Each polnt repre sents the current from indivtdual voitammograms integrated Over the range 400-800 mV YS. SCE. rne rewonse is ConverIed to concernration by calibration with dopamine. The solid llnes are the modeled response. Which involves the use of neurochemical kinetic p m m t e r s and diffusion l r m a dislance 01 10 pm. (b) A subtracted cyclic voltammogram obtained during the 60-Hz stimulation. The shape is Identical lo mat recwded in dopamine soiutions.

ANALYTICAL CHEMISTRY, VOL. BO. NO. 13, JULY 1. 1988

been given to support the identity of the released substance (29,30).The detected dopamine is almost 3 orders of magnitude lower in concentration than the concentration of HVA detected in the original experiments. This is because stimulus parameters of much lower intensity can he employed as a result of the higher sensitivity of the existing methods. Interpretation of the in vivo results The results in Figure 9 are satisfying at first glance-the goal of detecting dopamine concentration changes in the brain as a result of neural activity has been achieved. But, a t the same time, the answers to the last two analytical challenges listed on p. 112 A are not immediately obvious from inspection of the data. As noted earlier, the electrode is much larger than the synapse: therefore, the measurements must he from dopamine that “overflows” from the synaptic region into the extracellular space. However, the distance over which dopamine can diffuse during a 4-s, 30-Hz stimulation is small [l = (Dt)’/Z] and thus the dopamine must come from nerve terminals within an -50-pm radius of the electrode. The distribution of dopamine nerve terminals in the caudate nucleus is known to he heterogeneous, and we find that moving the electrode small distances inside the tissue can drastically alter the measured response, further supporting the view that we are sampling from a small, heterogeneous population of neurons (31). A second feature to note is that the dopamine concentration approaches a new steady-state concentration during the low-frequency stimulation. Thus the overflow of dopamine caused by the stimulation appears to he balanced by another process that removes dopamine from the extracellular fluid. The latter process is likely to be the uptake process described earlier. In fact, we find that the shape of the stimulation response can be altered hy administration of drugs that have been characterized by independent techniques to he competitive inhibitors of dopamine uptake (2831). We have described the observed response with a simple kinetic model based on the balance between uptake and stimulated overflow (31). However, to get a reasonable fit of the modeled data to that experimentally observed, we must convolute the modeled response with equations that describe diffusion over a distance of 1015 r m (32), as shown in Figure 10. These dimensions are similar to the radius of the electrode tip with its associated insulator and thus are reasonable on a physical basis. The diffusion process appears to give a time distortion of 200-400 ms between the observed response and that which occurred a t the

AFTER BEING 'ROUND FOR 13 YEARS,

WE'RE GOING STRAIGHT. Introducing J&Ws new Accusphere HPLC Columns -the

newest standard for your laboratory. For years, chemists have trusted our GC columns because they're so versatile. And durable. With high column-to-column reproducibility - the same features required in HPLC. Now, get that great J&W quality with Accusphere HPLC columns, available in three particle sizes and a wide range of bonded phases. Naturally, they come with J&Ws strong technical support and speedy delivery. See for yourself send for our free HPLC catalog. It's got data that'll set you straight. . . for J&Ws Accusphere. J&W Scientific, 91 Blue Ravine Road, Folsom, CA 95630, (916) 985-7888.

CIRCLE 78 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY. VOL. 60, NO. 13. JULY

I.

8yae

-

7 7 7 ~

I The Basics of Technical

synapse, and it also diminishes the amplitude of the observed response relative to that which occurred in the synaptic region. Quantitative in vivo measurements such as these are required so that models of the dynamics of dopamine neurotransmission can be tested. Recently, Justice published an extensive mathematical model of the nerve terminal region that can not only predict the type of behavior observed in this experiment but also a variety of other experimental conditions (33).The combined use of quantitative measurements and sophisticated models will allow a more complete understanding of the neurotransmission process. Future d i r e c t i i

C

ornrndnicat.ons %.iljmeyre a must for s.ccessf.l xentisu NOH YOL can ncrease pieffpctmess ano improve your commdntaoon skds Y th inis new easy t o - ~ s ereference b m Wfth I8 chapters. this hand) gdiae stam wlrh tne oas~s.sun as e.minaong %ord.ness ana jargon .n technical comrnunicauons. sing corren p-ncruation ana s e mn g approprm reios From there. p:i learn h m IO assembe yodr papers ana gather data for w i n (IImen ana oral repom Vw'd learn corrm oocument ng procmJres. inc .ding fwtnoting ana wmng a bibhog. raph). VOJ'II cover the .se of usual ads ana grapnrs. abnracr preparaoon. tne LSP of computers ana praofreaaing A woe banet) of praacal app Icat.ons IS COYered in this voI.me. inc .ding laborato? ana 0us.ness repom.,o.rnal puolcarons. granu ana proposals. o m e s s correspondence resmes. ana mema Tne &SICS of TKhflCd8 Commuxaong is a conbenieni. ewnnal ref. erence--one lnat )ou'II .se itme an0 1me again1 It tomp emenu Tne ACS Sryie GII& and W m g the IdWrdroo Norebwx io gw xienurn a comp.ete pxkdge for professional dwlopment in technical tornmmation SN.B ly 6. Edward Cain. Rocnester lnmrure of Technolcg ZOO pages ( 19881 ISBN 0-8412-1451-4 LC 88-3325

The success of the experiments described above lays a groundwork for the future. It is clear that the low levels of dopamine in the extracellular fluid are primarily a result of the potent uptake system that regulates dopamine concentration. This makes it unlikely that high concentrations of dopamine can diffuse from one synapse to another. However, this observation is for just one brain region, and it will be interesting to test whether this generalization occurs in other brain regions. Furthermore, we can only infer what is occurring in the synapse-smaller probes are required to test this. In addition, the response time is limited by the Nafion film. Therefore, more selective coatings are required with shorter transport times through the film. Although the other electroactive neurotransmitters can be detected with these electrodes, they have not yet been investi-

gated on the subsecond time scale. Much of the work to date has been done in anesthetized animals, and further work is required in unanesthetized animals. Furthermore, the sensitivity and selectivity needs to be improved so that the release of dopamine during the normal (as opposed to stimulated) firing of neurons can be measured. A whole new area of research inspired by Adams's original article is the measurement of events in single cells. Methods are being developed to measure the static and dynamic concentrations inside neurons. Microbore column chromatography has been used to separate the components of a single homogenized cell from an invertebrate (34). Microelectrodes are also being developed to probe dynamic changes from inside cells (35,36).These types of probes are required to see the part of the neurotransmission process that is being ignored with extracellular measurements. Finally, we note the point that A d a m made 12 years Go-an understanding of chemical communication between neurons does not mean that the ways in which the brain functions are fully understood. However, it should provide one additional clue in the overall attempt to understand the intricacies of brain function. Research in t h i s area has been supported by the National Institutes of Health and the National Science Foundation.

References (1) Adams. R. N. Anal. Chem. 1976.. 48.. 1126 A-1138 A. (2) Rose, G.; Gerhardt, G.; Strbmberg, I.; Olson,L.; Hoffer,B. Brorn Res. 1985.341, 92-100.

(3) Nicholson, C.; Phillips, J. M. J . Phy-

Distribution offlce Dept. 90

li55suteenthSt.tW Washington. D.C. 20336

01 CALL TOLL FREE

800-227-555t and use your credh card!

Flgure 10. Convolution of the modeled response. The changes in dopamine concentration predicted from a simple uplakelrelease model of dopamine dynamics in the exlraceIIuIar fluid are Shown by Curve A. Curve B illustrates the impulse response function associated with diffusion of m a m i n e from its site 01 release lo the electrode Surlace. When Cuwes A and 0 are convoluted with the Fourier transform approach, the result is Curve C.

778.4 * ANALYTICAL CHEMISTRY, VOL. 60, NO. 13, JULY 1. 1988

siol. 1981,321,225-57. (4)Ponchon, J.-L.; Cespuglio, R.; Gonon, F.; Jouvet, M.; Pujol, J. F. Anal. Chem. 1979,51.148346. (5) Fox, K.; Amstrong-James, M.; Millar, J. J. Neurosci. Methods 1980,3,37-48. (6) Kelly, R.; Wightman, R. M. Anal. Ckim. Acta 1986,187,7947. (7) Kovach, P. M.; Ewing, A. G.; Wilson, R.L.; Wightman, R. M. J. Neurosci. Methods 1984,10,215-27. (8) McCreery, R. L.; Dreiling, R.; Adams, R. N. Brain Res. 1974,73,23-33. Miller, E.; Rice, M.; Ad(9) Schenk, J. 0.; a m . R. N. Brain Res. 1983,277,l-8. (10) Huff,R.;Adams,R. N.;Rutledge,C. 0. Brain Res. 1979,173,369-72. (11) Gonon, F.; Buda, M.; Cespuglio, R.; Jouvet, M.; Pujol. J. F. Nature 1980,286,

____

m - nI d.. (12) Kovach, P. M.; Deakin, M. R.; Wight-

man, R.M. J. Phys. Chem. 1986, 90, 4612-17. (13) Gonon, F.; Fombadet, C. M.; Buda, M. J.; Pujol, J. F. Anal. Chem. 1981,53, 138689. (14)Wilson, R. L.;Wightman, R. M. Broin Res. 1985,339,219-26. (15) Gardiner, T.W.;Armstrong-James, M.;Caan,J. W.;Wightman,R. M.;Rehec, G . V. Brain Res. 1985,344,181-85. (16) Rebec, G. V.; Centore, J. M.; White, L. K.; Alloway, K. D. Science 1985.227, 438-40.

troanal. Ckem. 1985,188.85-94. (25) Kristensen, E. W.; Kuhr, W. G.; Wightman, R. M. Anal. Chem. 1987.59, 1752-57. (26) Miller, J.; Stamford, J. A,; Kruk, 2.L.; Wightman, R. M. Eur. J. Pharm. 1985, 109.34148. (27) Baur. J. E.; Kristensen, E. W.; May, L. J.: Wiedemann. D. J.: Wiehtman. R. M'. Anal. Chem., in press. (28) Kuhr, W. G.; Wightman, R. M. Brain Res. 1986,381,168-71. (29) Kuhr, W. G.; Ewing, A. G.; Caudill, W. L.: Wightman, R. M. J. Neurochem. 1984,43,560-69. (30) Ewing, A. G.; Wightman, R. M. J. Neurochem. 1984,43,570-77. (31) Wightman, R. M.; Amatore, C.; Engstrom, R. C.; Hale, P. D.; Kristensen, E. W.; Kuhr, W. G.; May, L. J. Neurosei., in press. (32) Engstrom. R.C.; Wightman, R. M.; Kristensen, E. W. Anal. Chem. 1988,60, 652-56. (33) Justice, J. B.; Nicolaysen, L. C.; Michael. A. C. J. Neurosci. Methods 1988.

-

(17) ,Neurochemical

Analysis of the Conscious Brain: Voltammetry and PushPull Perfusion; Annals of the NY Acade-

mv of Sciences: New York. 1987:D. 473. (W-Caliguri, E. J.; Meffoid, I. N. Brain

Res. 1984,296,15659. (19)Salamone, J. D.; Hamby, L. S.; Neill, D. E.; Justice, J. B. Pharmacol. Biochem. Bekau. 1984,20,609-12. (20) Wages, S. A.; Church, W. H.; Justice, J. B. Anal. Chem. 1986.58,1649-56. (21) Church, W.H.; Justice, J. B. Anal. Chem. 1987,59,712-16. (22) Church, W. H.; Justice, J. B.; Byrd, L. D. Eur. J. Pharm. 1987,34548. (23) Gerhardt, G. A.; Oke, A. F.; Nagy, G.; Moghaddam, B.; Adams, R. N. Brain Res. 1984,290,390-95. (24)Nagy, G.; Gerhardt, G. A.; Oke, A. F.; Rice, M.E.; Adams, R. N.; Moore, R.B.; Szentirmay, M. N.; Martin, C. R. J. Elec-

Suggested reading

more, 1986. Horn, A. S.; Korf, J. The Neurobiology of Dopamine; Academic Press: London, 1979. I

Hamamatsu Hollow Cathode Lamps are now available from major lab suppliers. Hamamatsu s ngle and multielement rlollow Cathode Lamps offer suDer or stability, spectral purity ana output intensity.even for sucn elements as arsenic and selendm They are compatible with most commercial spectrcphotometers. inc uding Beckman. Ze,ssand Perkin-Elmer And best of all. they re available from your local lab supplier

For Application Information,Call 1-800-524-0504

R. Mark Wightman (right)is o professor o f chemistry at Indiana University. H e received his B.A. f r o m Erskine College and his Ph.D. f r o m the University of North Carolina under the direction of Royce Murray. Following postdoctoral work with Ralph Adams at the University of Kansas, he joined t h e faculty of Indiana University. His research interests include microelectrodes and neurochemistry. Leslie J. M a y (center) is a graduate student in analytical chemistry at Indiana University. She received a B.A. degree in chemistry from t h e University of Kansas, where she did undergraduate research with Ralph Adams. Her interests include in vivo voltammetry and the voltammetry of biological compounds. Adrian C. Michael (left) is a postdoctoral fellow in chemistry at Indiana Uniuersity. H e received B.S. and Ph.D. degrees f r o m Emory Uniuersity. His research interests include i n vivo uoltammetry, applications of microelectrode arrays, and electrochemical investigations i n supercritical fluids.

1-201-231-0960 in New Jersey

HAMAMATSU HAMAMATSU CORPORATION

360 FOOTHILL ROAD P. 0. BOX 6910 BRIDGEWATER, NJ-7 PHONE: 201/231-w60

Q Hamamam rnomii 1886

CIRCLE 63 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL.

BO. NO. 13,

JULY 1, 1988

779A