Anal. Chem. 1994,66, 1468-1472
Temporal Response of Microdialysis Probes to Local Perfusion of Dopamine and Cocaine Followed with One-Minute Sampling A. P. Newton and J. B. Justlce, Jr.' Department of Chemistry, Emory University, Atlanta, Georgia 30322
The temporal response of microdialysis probes in vivo was followed with I-min sampling. The response under four conditionswas examined: in vitro sampling of dopamine (DA), in vitro perfusion of DA, in vivo perfusion of DA, and in vivo perfusion of cocaine. The temporal response of the probe in a stirred solution at 37 O C for bothsamplingmode and perfusion mode was determined using a flow rate of 1.6 pL/min. As expected, the rate of gain or loss of DA in these two cases rapidly stabilized. The in vivo response to a step change in perfusate concentration was also examined. The rate of loss of DA into the tissue rapidily reached a steady state so that the dialysate DA concentration stabilized within 2 min. To examine the in vivo response of DA to a step change in drug concentration, dialysis probes in the striata of anesthetized rats were perfused with artificial cerebrospinal fluid interchanged with 20 p M cocaine. The increase in DA caused by its uptake inhibition by cocaine rapidly reached a maximum level in 2-4 min and then gradually decreased. After the cocaine was removed, the dialysate concentration of DA fell rapidly with some tailing, suggesting the presence of residual cocaine in the tissue. These results suggest that relatively fast changes in DA levels in tissue can be followed by microdialysis probes. The rapid stabilizationalso suggests that the DA concentration profile does not extend very far from the probe. Microdialysis has become a well-established and flexible tool for the sampling of the extracellular fluid in the brain. Methods have been developed to determine quantitatively steady-state and transient extracellular concentrations of neurotransmitters such as dopamine (DA).'p2 The response of DA to many pharmacological agents is of particular interest, but because of relatively long sample intervals (10-20 min), partial loss in the temporal patterns may occur when relatively rapid fluctuations are taking place. In order to follow these fluctuations in concentration, a shorter sampling interval is necessary. A microdialysis sampling interval of 1 min for DA could be useful in several ways. First, as stated earlier, more rapid fluctuations in DA concentration could be followed. One application of this method is in drug self-administration studies in which animals self-administer drugs such as cocaine or amphetamine at intervals of 3-10 min. Much evidence indicates DA as a mediator for the reinforcing properties of (1) Parsons, L. H.; Justice, J. B., Jr. J. Neurochem. 1992.58, 212-218. (2) Olson, R. J.; Justice, J. B., Jr. Anal. Chem. 1993,65, 1017-1022.
M68
Ana&timI Chemistry, Vol. $8, No. 9, M y 1, 1994
drugs such as ~ o c a i n e ,and ~ ? ~therefore, the concentration of DA during self-administration is of interest. In addition, shorter experimental time for steady-state levels under different conditions would be possible. This in itself would be advantageous because the investigator is relieved of timeconsuming sampling, and less perturbation of the brain would occur when introducing exogenous substances for a shorter period of time. However, before one can meaningfully use a shorter time interval for following changes in concentration, the response of the sampling process to changes in concentration must be characterized. If the microdialysis process cannot respond to the changes occurring, then no benefit results from a shorter sampling period. The response time is largely determined by the time to establish a stable concentration gradient in the tissue surrounding the probe. The greater distance a substance must diffuse, the longer the time to establish steady state. Relatively inert materials, such as sucrose, and substances which are not rapidly metabolized, such as metabolites, require long periods to reach steady state around the probe.5 If the fluctuation in concentration of a substance is of interest, then a shorter sampling interval can only apply to transmitters and other rapidly metabolized materials. In the case of metabolites and inert materials, faster sampling would provide information about the time to reach steady state. One obvious potential problem in analyzing the small samples collectedis the detection limits for DAof the analytical system of choice. Dopamine has been determined to be present in low-nanomolar concentration in the nucleus accumbens and striatum of the rat brain under steady-state conditions'v6 using the point of nonet flux method.' The sensitivity available for DA through microbore high-performance liquid chromatography with electrochemicaldetection (HPLC-EC) has been suggestedto be sufficient for the analysis of 1-minmicrodialysis samples.* In analysis by HPLC-EC,DA is separated from interfering components and can be detected at very low concentrations (commonly 0.5 fmol per 0.5-pL injection, (3) Koob, G. F.; Vaccarino, F. J.; Almaric, M.; Swerdlow, N. R. In Cocuine:
Clinical and Biobehavioral Aspects; Fischer, S., Raskin, A., Uhlenhuth, E. H., Eds.; Oxford University Reas: New York, 1987; pp 80-108. (4) Wise, R. A.; Bozarth. M. A. Psychol. Rev. 1987, 94,469-492. ( 5 ) Morrison, P. F.; Bungay, P. M.; Hsiao, J. K.;Ball, B. A.; Mefford, I. N.; Dykstra, R. L.; Dedrick, R. L. In Microdialysis in the Neurarciences;Robinson, T. E., Justice, J. B., Jr., Eds.; Elscvier: New York, 1991; pp 47-80. (6) Smith, A. D.; Olson, R. J.; Justice, J. B., Jr. J. Neurosci. Methods 1992,44, 33-41. (7) Ltinnroth, P.; Jansson, P. A.; Smith, U. Am. J. Physiol. 1987, 253, E228E23 1. (8) Carlsson, A.; Sharp, T.; Zctterstrdm, T.; Ungcrstedt, U. J. Chromatogr. 1986, 36a,299-308.
0003-2700/~4/03651468~04.50/0
0 1994 Amerlcen Chemlcal Soclety
corresponding to 1.0 nM). Athough the amount of DA recovered by the probe in 1 min may not be sufficient to detect basal levels using HPLC-EC, behaviorally relevant changes occurring when DA is elevated may be followed. With improved detection limits, further applications of this method may be useful. Previously, 1-min sampling has been used to determine the temporal profiles of some exogenous substances introduced in high concentrations, such as ethanol? acetaminophen,lO and the antineoplastic, 3-amino- 1,2,4-benzotriazine 1,4-di-Noxide." Yet, endogenousneurotransmitters, such as DA, have yet to be sampled at this short of an interval. Part of the reason may be technical, but part of the reason may be the perception that the probe is sampling up to several hundred micrometers radially from the probe surface. Because diffusion over such a long distance would require a significant period of time, rapid sampling would not be useful for most applications of interest. In order to characterize the ability of the microdialysis process to follow changes that occur in the extracellular fluid of the brain, the probe response under four conditions was measured. First, the response to a step change in the sampled solution was measured. Second, the response was determined with respect to a step change in perfusion concentration. These in vitro experiments were performed in a stirred solution at 37 OC. Under these conditions, transport to the probe membrane was by convection rather than diffusion. The in vivo response to a step change in the perfusion DA concentration was determined and compared to the analogous in vitro experiment. The fourth experiment determined the response of DA to a step change in the perfusate concentration of a drug that increases the DA concentration. EXPERIMENTAL SECTION Probe Construction. Microdialysisprobes were constructed using a side-by- side design by inserting two different lengths of fused silica lines (40-pm i.d., 100-pm 0.d.; Polymicro Technologies, Inc.) into a 5-mm length of dialysis membrane (regenerated cellulose molecular weight cutoff 6000,220-pm 0.d.; Spectrum Medical) sealed at one end with polyimide resin (Alltech). The inflow and outflow lines were separated by 4 mm, which defined the active area. All other areas were sealed with polyimide resin. For all experiments, the probe was perfused from a 500-pL Hamilton syringe mounted on a Harvard Model 2274 syringe pump. In Vitro Response. To determine the response in sampling mode, the probe was perfused at 1.6 pL/min with artificial cerebrospinal fluid (aCSF) having the following composition: 145 mM NaC1,2.8 mM KC1,1.2 mM MgC12,1.2 mM CaC12, 0.25 mM ascorbic acid, 5.4 mM D-glucose, pH 7.2-7.4 (all chemicals obtained from Sigma). The probe was placed in a stirred aCSF solution at 37 OC for 5 min. The probe was then placed into a 40 nM DA solution for 15 min and back to the aCSF solution for 5 min. Samples were collected every minute in 250-pL microcentrifuge vials (Baxter Scientific) (9) Nurmi, M.; Kiianmaa, K.; Sinclair, J. D. Curr. Sep. 1993, 22, 46. (10) Chen, A.; Lunte, C. E.Curr. Sep. 1993, 12, 71. (11) Hogan, B.; Stobaugh, J. F.; Lunte, C. E.; Lunte. S. M. Curr. Sep. 1993, JZ, 81.
and stored on dry ice for analysis immediately after the experiment. For perfusion (delivery) mode, the probe was placed in a stirred aCSF solution at 37 OC. The perfusate was alternated between aCSF and 400 nM DA resulting in three 15-min perfusions with aCSF and two 15-min perfusions with 400 nM DA. Sample collection was carried out as above. The first five samples after each change in perfusion medium were analyzed in order to see the early response. For the remaining 10 min, every other sample was analyzed. In Vivo Response to DA and Cocaine Perfusion. Male Wistar rats (300-330 g; Harlan Sprague-Dawley Inc.) were anesthetized with 400 mg/kg chloral hydrate and placed in a stereotaxic apparatus. The probe was placed in the striatum using coordinates AP +2.7, MS -2.7 from bregma and DV -6.7 from dura; incisor bar at +5.0 mm.12 Baseline samples were collected at 1.6 pL/min after an equilibration period of 90 min. For the in vivo system response to perfusion of 400 nM DA, the perfusate was alternated between aCSF and 400 nM DA for a total of three 15-min perfusions with aCSF and two 15-min perfusions with 400 nM DA. Samples were collected every minute in 250-pL microcentrifuge vials and stored on dry ice for analysis immediately after the experiment. The first five samples after each change in perfusion medium were analyzed in order to see the early response. For the remaining 10 min, every other sample was analyzed. A concentration of 400 nM DA was used because the loss of DA during perfusion with high concentration of DA, such as 400 nM, is being used to characterize DA uptake.13 Knowledge of probe response is therefore needed under these conditions. For the response to cocaine, the perfusate was interchanged between aCSF and 20 pM cocaine resulting in three 15-min perfusions with aCSF (including baseline collection) and two 15-min perfusions with the cocaine solution. Samples were collected and analyzed as in the previous experiment. Chromatographic Conditions. Samples were analyzed for DA concentration by microbore HPLC with electrochemical detection using a 0.5 mm X 10 cm stainless steel column with 5-pm C6 stationary phase. Either 0.5 or 1.OpL was injected. The mobile phase consisted of 27.2 mM sodium phosphate buffer with 15% methanol (v/v), 0.13 mM disodium ethylenediaminetetraacetate (Na2EDTA), 1 mM octyl sodium sulfate, pH 3.7 (all chemicals obtained from Sigma). An Isco LC-5000 syringe pump delivered the mobile phase at a flow rate of 17 pL/min. Dopamine was detected with either an EG&G Princeton Applied Research (Model 400) or a Bioanalytical System (Model LC-4C) amperometric detector using a working and a reference electrode (Model RE1) from Bioanalytical Systems Inc. The applied potential was +0.7 V versus Ag/AgCI. Samples were loaded manually with a 10-pL Hamilton syringe. The loop was rinsed with distilled water and dried with a 1-mL syringe after each injection. A calibration curve consisting of 5,10, and 20 nM DA was used to quantitate the dialysates for the in vivo experiments involving the cocaine; 100,200, and 400 nM DA for the perfusion mode in vivo and in vitro responses; and 10,20, and 40 nM DA for the sampling mode in vitro response experiment. (12) Pellegrino, L.;Pellcgrino, A.; Cushman, A. A Stereotaxic Atlas of the Rat Bruin, 2nd ed.; Plenum Press: New York, 1979. (13) Smith, A. D.; Justice, J. B.,Jr., submitted for publication in J. Neurochem.
Analytical Chemistry, Vol. 66, No. 9, May 1, 1994
1469
i
300
-
250
-
200
-
lo
sc
-
C
0; C
0
0
150
83
100
,
.
-E .g
n O C
.m
40 nM DA I
I
-15
0
15
:
0
aCSF
I
50.
J
-Zs+
b*=-=eCSF
400nMDA I
I
I
30
Time (min) Figure 1. In vitro microdialysis response (1 trial) in a stirred solution at 37 OC. A 4-mm probe was transferred from a 0 nM dopamine solution to a 40 nM dopamine solution. Samples were taken every minute using a dialysate flow rate of 1.6 pL/min and analyzed for dopamine concentration on microbore HPLC with electrochemical detection. A maximum response occurs within 2 min.
Histology. At the end of the experiment the anaesthetized animals were perfused with saline followed by 10% formalin. The brain was removed and stored in 10% formalin. To verify placement of the probe, the brain was frozen on a cryostat and sliced into 50-pm sections which were stained with thionine. Slices were microscopically examined and compared to an atlas.I2 RESULTS AND DISCUSSION The purposes of this study were to characterize the temporal response of microdialysis probes in vivo and to test the feasibility of microdialysis with 1-min sampling for DA using microbore HPLC-EC as the means of detection. In order to overcome potential difficulties of sampling and detection, conditions of these initial experiments were optimized. Short outflow lines were used to reduce back pressure and dispersion of the sample due to longitudinal diffusion. Anesthetized animals were used. A probe with a 4-mm active area for use in thestriatum (as opposed to 2 mm for the nucleus accumbens) was chosen for increased recovery. Use of the smaller active area is not foreseen as a problem when changes in elevated levels of DA such as in cocaine or amphetamine selfadministration are being examined. The flow rate of 1.6 p L / min was chosen for the minimum volume that could be manually handled and loaded onto the HPLC sample loop. InVitroResponse. Theinvitro experimentswere performed in order to determine the fastest possible probe response and to compare these to analogous in vivo experiments. Figure 1 shows one trial of the in vitro response for a stirred solution at 37 OC with the probe operating in sampling mode, expressed as concentration of dialysate (nM DA) versus time. After the solution being sampled is changed from aCSF to 40 nM 1470
Analyticel Chemistty, Vol. 66, No. 9,h&y 1, 1994
DA, the dialysate reaches a stable level within 2 min. Upon switching back to aCSF, the residual DA is removed within 2 min also. Figure 2 contains the in vitro response upon switching the perfusate from aCSF to 400 nM DA while sampling from an aCSF solution. The data represent the mean f SEM for a total of four perfusions in two probes (two perfusions per probe). As in the previous case, the dialysate reaches a stable level within 2 min of changing the perfusate concentration. As seen from Figures 1 and 2,the response in vitro is very rapid, particularly when the 1 5 s lag time from the tubing volume is considered. After switchingback to initial conditions (aCSF being sampled for the sampling mode and aCSF in the perfusate for the perfusion experiment), the dialysate concentration returns to baseline as rapidly as it rose. The data also show that there is somevariation in the measureddialysate concentration for the samples collected during steady state. This appears to be due, in part, to the analytical measurement of such small samples. Comparison of peak heights obtained from loading 1.2 pL through a 0.5-pL loop (typical for these experiments) to loading the same loop with 100 p L shows a variation over 5 times larger for the smaller load volume (SEM of 2.6% and 0.48%, respectively, for N = 10 injections, data not shown). The relative recoveries, or extraction fractions (nM DA gainedf40 nM for sampling experiment; nM DA lost/400 nM for perfusion experiment), for the sampling and reverse modes were 31.7 f 0.6% and 34.6 f 1.3% (mean f SEM), respectively, and are not significantly different (p > 0.05), demonstrating the symmetry of the microdialysis process. These perfusion data show that the in vitro response under these conditions is fast, as expected. In a stirred solution, mass transport through the surrounding medium is primarily
by convection rather than diffusion and is therefore rapid. The membrane of the probe in this case becomes the ratelimiting step in probe response.14 However, the choice of sampling interval limits the resolution of the measured response of the microdialysis probe. From these data, the response time is as fast or faster than that observed. In Vivo Response to Dopamine Perfusion. It is not possible to directly make a step change in extracellular DA concentration. However, it is possible to make a step change in the concentration of DA in the perfusion fluid to levels significantly above the extracellular level of about 10 nM6 and follow the time for the rate of loss to the tissue to reach steady state. Figure 2 compares the in vitro and in vivo responses when the probe supplies material to the surrounding medium by switching the perfusate from aCSF to 400 nM DA. The time for steady state to be reached in vivo as determined by loss of DA from the probe to the tissue is just as short as the analogous in vitro experiment. That is, a steady-state concentration of the dialysate is reached by the second sample after the perfusate is switched to 400 nM DA and also upon return to the aCSF solution for both the in vivo and in vitro systems. Note that the time to steady state in vivo following a step to 400 nM DA includes the time needed to restore the depletion region created by perfusing with aCSF and establish a stable rate of addition of DA to the tissue. Similarly, the time to steady state after switching back to aCSF includes the elimination of the high concentration in the extracellular fluid adjacent to the probe and the creation of a depletion region. The linearity of the amount of DA gained or lost versus the amount being perfused has been demonstrated for a variety of conditions,15so that although DA loss to tissue is measured in the present in vivo experiment, a similar time course should exist, as was shown in vitro, for the situation in which DA is being sampled from the tissue. These data indicate that the in vivo response to local perfusion with DA is rapid and that the gradient in the surrounding tissue rapidly stabilizes, which can only happen if DA does not diffuse very far into the tissue. The DA concentration gradient at the surface of the microdialysisprobe appears to stabilize within a few minutes, resulting in a steady state of mass transport of DA to or from the probe. This result is contrary to what is expected for cylindrical diffusion, which shows a log dependence on time.16 Steady state is not achieved in the absence of additional contribution to the flux. The rapid stabilization of response for DA results from the high density of DA nerve terminals in the tissue surrounding the probe, which release the transmitter into the extracellular fluid and remove it via reuptake. These active processes significantly affect probe responsesJ5 from what it would be if only passive diffusion through the tissue were occurring. Removal of the DA nerve terminals by 6-hydroxydopamine (6-OHDA) lesions significantly reduces the flux of DA to or from the probe. The reduced flux, as measured by the significantly reduced extraction fraction, occurs in the absence of a 6-OHDA lesion effect on the extracellular concentration.17 (14) Bungay, P. M.; Morrison, P. F.; Dcdrick, R. L.LifeSci. 1990,46,105-119. (15) Justice, J. B., Jr. J . Neurosci. Merhods 1993,18, 263-276. (16) Amatore, C. In UIrramicroelecrrodes;Flcischmann, M.,Pons, S.,Rolison, D., Schmidt, P., Us.; Datatech Systems, Inc.: Morganton, NC, 1987; p 81. (17) Parsons, L. H.;Smith, A. D.; Justice, J. B., Jr. J. Neurosci. Methods 1991, 40. 139-147.
Removal of the DA nerve terminals reduces the ability of the tissue surrounding the probe to supply DA to the probe when it is operated in the sampling mode or remove DA when the probe is operated in the delivery mode. Taking the effect of the DA nerve terminals into account, a sampling distance of less than 50 pm has been estimated for DA when the perfusate flow rate is 0.6 pL/min.l* This distance indicates the importance of active processes to the flux of material to or from the probe. It also indicates that the probe is not sampling the transmitter from other neuronal regions hundreds of microns away from the probe. In contrast to DA and other neurotransmitters, the behavior of substances that have much slower dynamics is very different. Neurotransmitters have a very high turnover in the extracellular fluid. Metabolites, such as dihydroxyphenylacetic acid (DOPAC), have lower rates of extracellular turnover and therefore have concentration gradients that extend farther out from the probe than neurotransmitters. The distance that the probe exerts an effect on the surrounding concentration is strongly influenced by the rates of processes that supply and remove the given material. The slower the kinetics of these processes, the greater the distance of the probe's effects, and the longer the time to reach steady state. For a probe operated at 1 pL/min, the distance from the probe surface to the point at which the DOPAC concentration is half the undisturbed extracellular concentration has been estimated to be 0.2 mm.5J4 The relatively inert substance sucrose produces a concentration gradient that extends considerably farther than DOPAC. For sucrose the equivalent distance has been calculated to be 3 mm. Both of these compounds would be expected to take considerably longer to reach steady state than DA. In Vivo Response to Cocaine Perfusion. Panels a and b of Figure 3 show the increase in dialysate DA concentration obtained in vivo when 20 pM cocaine is added to the perfusate. Figure 3a expresses the data as dialysate DA concentration. Figure 3b shows the data as percent of the maximum dialysate concentration. This removes the interanimal variation and demonstrates that the shapes of the curves obtained for each animal are very similar. Although the shapes of the two figures are the same, much more variation occurs in the data expressed as concentration. A concentration of 20 pM cocaine was used to mimic selfadministration conditions. A typical mean drug intake rate during cocaine self-administration is 0.3 mg/kg per minute.19 The in vivo concentration of cocaine in the rat brain for this same perfusion dose administered as a constant iv perfusion was determined to be 17.1 pM.20 The levels of DA measured in the present workshould therefore be similar to those achieved during cocaine self-administration. Thus, the present results support the feasibility of the method for monitoring DA during cocaine self-administration. Dopamine reaches a maximum concentration early (2-4 min) during the first 15-min cocaine perfusion and then slowly declines. The second cocaineperfusion results in a significantly lower maximum (p < 0.005). After the cocaine is removed (18) Bungay, P. M.; Smith, A. D.; Olson-Cosford,R. J.; Newton, P.; Justice, J. B., Jr. Soc. Neurosci. Absrr. 1993, 19, 1848. (19) Pettit, H. 0.;Justice, J. B., Jr. Brain Res. 1991, 539, 94-102. (20) Mcnacherry, S.;Hubert, W.;Justice, J. B., Jr. AMI. Chem. 1992,64, 577583.
AnalyticalChemistry, Vol. 66, No. 9, M y 1, 1994
1471
N=4
1
100
C 0:
--r-in
-
9
80
8 E
z
60-
.-
X
9
40-
CI
-15
0
15
30
45
60
Time (min)
5
g a
I
I
20-
.e' AA A A 0-
^%
J
LA
N=4
0
vivo in vitro
A
4
caxd
A A A A A
~
oox)
I
I
I
I
-15
0
15
30
Time (min) Flgurr 4. Flgures 1 and 3b combined to give a comparison of the corresponding In vttro and in vivo responses expressed as percent of the maxlmum concentratlon. Error bars were deleted from Figure 3b for clarity.
20&4Mine
CSF -15
0
15
CSF
I
I
30
45
eo
Time (min) Flgurr 3. Mlcrodialysis wlth 1-min-sampling in the striatum expressed as (a) dopamine dialysate concentratlon and (b) percent maxlmum (meen f SEM; N= 4). The perfusate medlum was interchangedbetween aCSF and 20 pM cpcalne, resuklng In three lbmln perfusions wlth aCSF and two 15-min perfusions with cocaine In each anlmal.
from the perfusate, the DA levels drop rapidly yet do not reach baseline levels for 5-7 min. Figure 4 compares the in vitro response from a step change in the sampled DA solution concentration to the in vivo probe response from the cocaine perfusion. The data are expressed as percent of the maximum dialysate DA concentration, and the error bars have been removed for clarity. Figure 4 shows that the in vivo response is rapid and that the decline in vivo after removal of cocaine is relatively slow compared to that in vitro. There are at least three factors that affect the response of the probe to the cocaine perfusion. The first is the time for the drug to diffuse into the tissue. Although this process is apparently fast due to the relative lipophilicity and rapid metabolism of cocaine,21it is probably the slowest step in the process. Next is the time it takes for the drug to increase the DA concentration by binding to the DA transporter. This is likely to be very fast on the present time scale. The third limitation to response is the time it takes for the DA to diffuse back to the probe. This should be similar to the time for diffusion of DA from the probe to the tissue as in Figure 2. The overall process is evidently rapid as the signal maximizes within 2-4 min. Interestingly, for thesecond cocaineperfusion, (21) Pan,H.; Mtnacherry, S.; Justice, J. B., Jr. J. Neurochem. 1991, 56, 1299-
Analytcal Chemistry, Vol. 66,
CONCLUSION In summary, a sampling interval of 1 min for microdialysis is feasible under the conditions described. The rapid changes observed and the short response time of the sampling process indicate that the microdialysis probe is capable of following rapidly fluctuating levels of DA. The rapid response in vivo also suggests that DA is sampled from very close to the probe surface. These results indicate that microdialysismay be used for studies in the neurochemistry of behavior where slower sampling is inadequate. The present method may allow monitoring of the DA fluctuations with short enough sampling intervals to better relate neurochemistry to behavior. ACKNOWLEDGMENT Support from NSF Grant BNS-9111617 is gratefully acknowledged. Recelved for review December 14, 1993. Accepted February 15, 1994." _____
_ _ _ _ _ _ ~
~~
Abstract published in Advance ACS Absrracts, April 1 , 1994.
1306.
1472
the response is as fast, but the levels are not as high as those of the first cocaine perfusion. The decline in DA after cocaine is removed from the perfusate is also rapid, yet it does not reach baseline levels for 5-7 min. This tailing is probably a result of residual cocaine in the tissue. The decline that occurs during the first cocaine perfusion may be due to several causes, including decreased release and a decrease in the concentration gradient around the probe as uptake is inhibited. Reduced uptake is known to reduce probe recovery.* For this paper, the focus is the microdialysis sampling process itself rather than the interpretation of the effect of cocaine.
NO.'^, May
1, 1994
