Journal 01Internal Medicine 1991; 2 3 0 : 365-373
Microdialysis-principles animals and man
and applications for studies in
U. UNGERSTEDT From the Department o/ Pharmacology. Karolinska institute. Stockholm. Sweden
Abstract. Ungerstedt U (Department of Pharmacology, Karolinska Institute, Stockholm. Sweden), Microdialysis-principles and applications for studies in animals and man. Journal of lnternal Medicine 1991 : 230: 365-373
Microdialysis is a technique for sampling the chemistry of the individual tissues and organs of the body, and is applicable to both animal and human studies. The basic principle is to mimic the function of a capillary blood vessel by perfusing a thin dialysis tube implanted into the tissue with a physiological liquid. The perfusate is analysed chemically and reflects the composition of the extracellular fluid with time due to the diffusion of substances back and forth over the membrane. Microdialysis is thus a technique whereby substances may be both recovered from and supplied to a tissue. The most important features of microdialysis are as follows: it samples the extracellular fluid, which is the origin of all blood chemistry; it samples continuously for hours or days without withdrawing blood ; and it purifies the sample and simplifies chemical analysis by excluding large molecules from the perfusate. However, the latter feature renders the technique unsuitable for sampling large molecules such as proteins. The technique has been extensively used in the neurosciences to monitor neurotransmitter release, and is now finding application in monitoring of the chemistry of peripheral tissues in both animal and human studies. Keywords : extracellular fluid, microdialysis.
Introduction In medical practice, most sampling from the body is achieved by drawing blood. It is relatively straightforward, acceptable, and the reference ranges are well known. However, blood merely reflects the chemical composition of the extracellular fluid of the individual organs. In many instances it would have been more relevant and informative to sample this fluid directly. Due to the development of a new technique, microdialysis, this is now possible. Most experimental attempts to sample from the extracellular environment have been performed on the brain. Gaddum [l] devised a ‘push-pull’ technique whereby liquid was pumped (pushed) directly into the brain tissue through one cannula, while it was simultaneously withdrawn (pulled) from the tissue at the same speed by an adjacent cannula. However, the push-pull technique suffers from the drawback that the tissue is disturbed by the streaming
liquid-it is necessary to balance the push and pull flows exactly in order to avoid sucking out tissue or injecting liquid. Bit0 et al. [2] were the first to implant ‘dialysis sacs ’ containing 6% dextran in saline into the subcutaneous tissue of the neck and the parenchyma of the cerebral hemispheres of dogs. Several weeks later they removed the sacs surgically and analysed their amino acid content. These studies introduced the concept of a ‘compartment’ surrounded by a dialysis membrane which equilibrates with the extracellular environment. However, the intention was not to monitor changes over time, but to avoid physiological fluctuations and to obtain an average concentration of the substances of interest. Delgado et al. [ 3 ] developed a ‘dialytrode’ consisting of two stainless steel tubes forming a pushpull cannula ending in a small permeable bag made of dialysis membrane. The dialytrode was tested in vivo in monkey brain tissue. 365
366
U. UNGERSTEDT
In our own laboratory we implanted ‘hollow fibres’ into the brain in order to mimic the function of blood vessels. When dopamine neurones were labelled by perfusion with 3H-dopamine we could monitor baseline and amphetamine-stimulated release in anaesthetized and freely moving animals in relation to behavioural activation [4]. The simple hollow fibre ‘ microdialysis probe ’ has subsequently been extensively improved, so that it is now possible to implant a ‘needle probe’ (Fig. 1) with a hollow fibre membrane at its tip [5-81 into brain and other tissues in much the same manner as a cannula or an electrode. We have applied this probe to studies on subcutaneous [9], brain [lo, 111 and muscle tissue [12] in man.
Features of microdialysis The basic principle of microdialysis is to mimic the passive function of a capillary blood vessel by perfusing a thin dialysis tube implanted into the tissue. The perfusate is analysed chemically and reflects the composition of the extracellular fluid over time due to the diffusion of substances back and forth over the membrane. Microdialysis is thus a technique whereby substances may be both recovered from and administered to a tissue. The most important features of microdialysis are as follows. (i) It samples the extracellular fluid as distinct from the whole tissue collected by biopsies, punches or dissection. (ii) It can be performed locally in almost every organ and tissue of the body (including blood). (iii) It makes it possible to sample continuously for hours or days. (iv) It can be used to recover and/or introduce endogenous and exogenous substances in the tissue. For example, it is possible to administer a drug systemically while using microdialysis to estimate the local drug concentration, the local biochemical effect and the resulting physiological response. (v) It collects a representative sample of all substances in the extracellular fluid (provided that they pass across the membrane), carries them out of the body and makes them accessible to conventional analytical techniques. This distinguishes the technique from in-vivo sensor methods, such as implantable biosensors. (vi) It causes minimal damage to the blood-brain barrier [8, 131. This makes it possible (a) to
study drug penetration into the brain by analysing the drug concentration in the perfusate and (b) to compare the effects of drugs on the brain when applied directly via the microdialysis probe or via the systemic route. (vii) It simplifies perfusions. The flow is unidirectional, and there is no need to balance the flow as in a push-pull cannula. (viii) It minimizes damage to the tissue, as there is no direct contact between the fluid flowing outside the membrane and the cells of the tissue. (ix) The membrane prevents large molecules from diffusing into the perfusate. This protects certain molecules from degradation by tissue enzymes once they have been recovered from the tissue. (x)The process of microdialysis purifies the sample, so that it is possible to introduce it directly into the analyticai instrument, for example a n HPLC system. (xi) It introduces a sterility barrier between the perfusion liquid and the tissue. (xii) The size of the perfused area can be regulated by varying the length of the membrane. (xiii) Microdialysis can be performed in blood or in body cavities such as the uterus, the peritoneal cavity and the mouth. This is impossible with an open push-pull system. (xiv) The performance of a microdialysis probe and its membrane can be studied in vitro. Different probes can be compared with regard to the recovery of a particular substance, or a single probe can be studied for its ability to recover substances of different molecular weight or charge. (xv) Clinically, microdialysis allows repeated sampling for hours or days after a single penetration of the tissue, without the need to withdraw or handle any blood.
Principles of microdialysis Microdialysis is a very simple technique. A tubular dialysis membrane is introduced into the tissue or placed in contact with a moist surface, for example a mucous membrane. The tube is perfused with a liquid that equilibrates with the fluid outside the tube by diffusion in both directions. The degree of equilibration is subject to known laws of physical chemistry. The complexity of the technique is due to the interaction between the dialysis tube, the perfusion
PRINCIPLES OF MICRODIALYSIS
0
.O 0 0 '
0
0
0
.
0
0 0
Fig. 1. Microdialysis probe. The perfusion liquid enters through the right inlet cannula and passes through an inner cannula all the way to the tip of the probe (see enlargement of the tip) where it flows upwards between the inner cannula and the membrane, where dialysis takes place. The liquid leaves the probe through the left outlet cannula.
liquid and the living tissue. In order to perform microdialysis and to interpret the results correctly, it is important to envisage the many events that take place during this interaction. There is a rapid fall in the concentration of most substances in the perfusate during the first phase of microdialysis. which is probably due not only to an initial lesion of the tissue which causes excessive release of substances from cellular storage compartments, but also to the establishment of a new steady-state level of most extracellular substances because of drainage through the probe [14. 151. In our experience the degree of initial damage determines how rapidly baseline levels are reached, and care should be taken to introduce the probe slowly into the tissue. During the initial period of disturbed tissue function glucose metabolism is increased, blood flow decreased, and transmitter release is disrupted. This period lasts from minutes to hours, depending on the
367
tissue, the functions of which are measured and the substances analysed [16-181. Baseline levels are reached more rapidly in larger organs where the relative size of the probe is smaller. In the human brain this often occurs within 10-20 min [Ill. A related problem is the time period over which substances can be reliably recovered once they have reached 'baseline ' levels. This varies between organs, substances and investigators [18. 191. Such variation is not surprising in view of the biological differences between tissues with regard to synthesis, storage, metabolism, density of innervation, delicacy of nerve terminals, regional difference in blood flow, etc. There are also obvious differences between laboratories and the types of probe employed, use of guide cannula, maintenance of sterility, implantation procedures, composition of the perfusion medium, biocompatibility of membrane, species, etc. In clinical experiments we have been able to maintain constant levels of various metabolites in the subcutaneous tissue for 6-7 d (unpublished data). The microdialysis probe communicates with the extracellular fluid by diffusion along a logarithmic concentration gradient towards and away from the probe. The direction of this gradient is dependent on the composition of the perfusion liquid, and the latter dominates the immediate environment of the probe. If calcium is excluded from the perfusion liquid, the extracellular fluid surrounding the probe is depleted of calcium, impairing synaptic transmission [20, 2 11. Conversely, inclusion of a substance in the perfusate will make it spread from the probe along a concentration gradient the direction of which is opposite to that of calcium depletion. The composition of the perfusate should be as close as possible to normal physiological levels of the most essential compounds in the extracellular environment [22]. The extent of this similarity must be determined for individual cases, as the aim of a microdialysis experiment is to remove or administer substances by exploiting the differeqres between the perfusion liquid and the extracellular fluid. The recovery of substances from the extracellular fluid is dependent on (i) the length of the dialysis membrane, (ii) the flow of the perfusion liquid, (iii) the speed of diffusion of the substance through the extracellular fluid and (iv) the properties of the membrane [15, 14, 23, 241. The speed of diffusion is a property related to diffusion through the parenchyma as well as to active elimination from the tissue by uptake into cells and blood capillaries. The
368
U. UNGERSTEDT
recovery is the same in both directions across the membrane [15]. For small molecules, such as the monoamine transmitters, the factor that limits recovery is usually the speed of diffusion through the extracellular fluid, and not diffusion through the membrane [ 151. In the tortuous brain parenchyma the speed of diffusion may be around 50% of that in water, while the corresponding figure for the membrane may be as high as 80%. Although it is tempting to think that the membrane represents an important barrier, it is actually the ability of a substance to diffuse through the extracellular space that determines both the amount of substance that can be recovered and the speed with which a change can be detected. It is convenient to distinguish between relative recovery (concentration recovery) and absolute recovery (mass recovery) [7]. Relative recovery is the concentration of a particular substance in the perfusate when it leaves the probe, expressed as a percentage of the concentration in the surrounding medium. Absolute recovery is the total amount of substance recovered during a defined time period, expressed in mol I-'. When a microdialysis probe is tested in vitro, relative recovery decreases and absolute recovery increases as the perfusion flow is increased. Both values reach a plateau when the diffusion speed through the surrounding medium reaches its maximum. In vivo the relative recovery is constant as long as the perfusion conditions remain the same. However, the absolute recovery of a substance varies with its production/release in the tissue. The usual reason for performing microdialysis is to monitor this change in the extracellular levels of endogenous or exogenous substances.
Neurotransmitter release A central question is whether recovered transmitters reflect ' true ' synaptic release or a more unspecific overflow from synaptic and non-synaptic sources. Blocking of the dopamine [25]. 5-HT [26] or noradrenaline [2 71 reuptake mechanisms increases the level of the transmitters in the perfusate, and stimulation of dopamine autoreceptors [2 81 gives the expected decrease in dopamine. Perfusion with TTX, in order to block sodium channels, lowers acetylcholine [29], dopamine [191 and noradrenaline [2 71 levels. The omission of calcium lowers dopamine [20], 5-HT [26] and acetylcholine [30] levels.
Although these data strongly suggest that the microdialysis probe recovers an overflow from the synaptic release, they do not prove that this overflow is quantitatively related to synaptic release. Studies of GABA are more complicated. Uptake block increases GABA in the perfusate [31], while the effect of low calcium and TTX varies according to the smoothness of the implantation procedure, the time period between implantation of the probe and the experiment, and whether the animal is anaesthetized or conscious. Drew et al. [17] and Westerink and De Vries [25] did not find any effects of TTX in acutely implanted, anaesthetized animals, while Osborne et al. [18] were able to demonstrate decreased GABA after both TTX and omission of calcium in conscious animals in which the microdialysis probes had been implanted for 1 4 d. The correct interpretation of the GABA experiments is probably that GABA in the perfusate does not originate solely from synaptic sources. However, the fact that GABA in the perfusate is increased by dopamine D1 receptor agonists and lowered by D2 agonists [32] seems to provide positive evidence for release under synaptic control.
Finding the ' true ' extracellular concentration Another fundamental problem is the relationship between the concentration in the perfusate and that in the extracellular fluid. Several attempts have been made to determine this concentration by extrapolation from in-vitro recovery values, but there is now general agreement that this is not reliable because of the difference in diffusion coefficients between water and tissue. The best estimates of extracellular concentrations are obtained from in-vivo experiments. They can be calculated by varying the perfusion flow during a n in-vivo experiment, measuring the change in the substance of interest coming out of the probe, and then extrapolating to zero flow [33], or by perfusing with varying concentrations of the substance and then calculating the equilibrium concentration, i.e. the concentration at which the substance in the perfusate does not change during the perfusion because it has the same concentration inside the probe as in the extracellular fluid [34]. Both these invivo methods require that the concentration in the extracellular fluid remains constant throughout the experiment.
PRINCIPLES O F MICRODJALYSIS
A third method is to use a reference substance in the perfusate, or to perfuse with it before the systemic injection of an exogenous compound (StBhle. to be published). The method is based on the fact that recovery over the membrane is the same in both directions. Given that the diffusion properties of the reference substance in the perfusate and the substance to be recovered from the brain are the same, the percentage loss of the reference substance from the perfusate will be the same as the percentage recovery of the substance from the brain. This recovery value is then used to calculate the concentration in the extracellular fluid. Thus in pharmacokinetic experiments the true extracellular concentration of a drug can be determined by equilibrium dialysis ' during steady-state conditions, or by including a reference substance in the perfusate. This substance should have the same diffusion characteristics as the systemically administered drug, ideally appear as a peak in the same chromatographic system, and be pharmacologically inactive at the concentration used.
The microdialysis probe The reason for making a microdialysis probe is to provide a convenient means of introducing a dialysis tube into the tissue. There are basically three types of probe : the dialysis tube per se, the loop probe and the concentric probe. The advantage of using the dialysis tube per se is that it is simple to make such a probe. The drawback is that it is necessary to make holes for both entry and exit. The loop probe consists of two parallel metal tubes, connected by a dialysis tube forming a loop [6]. In order to stretch the loop during implantation into the tissue, a stylus is placed inside or alongside one of the metal tubes, reaching all the way to the distal end of the loop. The loop probe is easy to make and position in a tissue. However, the two parallel tubes increase its size and thus enhance the damage to the tissue. In the concentric probe [8] a piece of dialysis tubing is sealed by glue at one end forming the tip of the probe (Fig. 1).The other end is usually glued into a tube which is the shaft of the probe. A thin inner cannula extends through the shaft and dialysis tube all the way to the tip. The material may be metal, fused silica or plastic. The perfusing liquid usually enters the proximal end of the inner cannula and flows distally all the way to its end, where it changes
369
direction and returns in the space between the inner cannula extends and the membrane where microdialysis takes place. The proximal end is designed in such a way that the inlet and outlet are separated. The concentric probe is the most difficult to construct. It can be made very thin, and the length of the tip can be varied from mm to cm in length. The perfusion liquid is carried to and from the probe by thin tubes, the importance of which is often neglected. They usually represent a substantial inner volume which delays the introduction of substances into the probe, as well as collection of the samples. When using a low perfusion flow rate it may be many minutes from the time the drug is included in the perfusate until it reaches the probe in the tissue, and still longer before it reaches the collecting vial. The material of the tubing should be as inert as possible in order to avoid adsorption of substances to its walls. The perfusion flow rate should be low (0.1-5 pl min-') in order to remove as little as possible of the extracellular content and thus minimize the interference with normal physiology. This demand must be balanced against the difficulty encountered in handling very small amounts of fluid and the need to obtain enough material for the analytical technique. In large homogenous organs such as the liver or adipose tissue, it is possible to compensate for a low flow by increasing the length of the membrane. Under ideal conditions, e.g. during microdialysis in the blood, the outer environment remains constant due to the streaming blood. With a 20-mm membrane in a vein of a pig the recovery of lactate is virtually 100% at a flow rate of 2-3 1 1 min-'.
Microdialysis instruments The technical difficulties of microdialysis relate to the handling of low flow rates and small volumes. However, most syringe pumps on the market can produce the required flow rates. A few instruments have been developed specifically for microdialysis, such as the CMA 100 series, consisting of a syringe pump which powers and controls a syringe selector, a microfraction collector or an injector for on-line injections into a chromatograph. The CMA 200 series includes a refrigerated fraction collector in which the sample cassettes can be transferred directly to a refrigerated autosampler and the samples injected into one or two different chromatographs.
370
U. UNGEKSTEDT
Fig. 2. Schematic drawing of the instruments used during a microdialysis experiment on a rat. A = computer controlling the experiment, B = perfusion pump with three syringes containing different perfusion fluids, C = syringe selector which selects one of the syringes of the pump (B) under computer control (A), D = anaesthetixed rat in a stereotaxic instrument. The microdialysis probe is held in the instrument and placed in the brain, E = microfraction collector, which collects fractions under the control of the computer (A) or the pump (B). F = second pump used to inject a drug systemically under the control of the computer (A), G = temperature controller to maintain the anaesthetized animal at the correct temperature.
The use of a syringe selector is important if the introduction of air bubbles into the perfusate, which often occurs if the tubing is transferred manually between syringes when changing the perfusion liquid (Fig. 2). is to be avoided. A stereotaxic instrument is necessary for microdialysis studies on the brain. It is a great advantage if the stereotaxic implantations can be performed under a stereo microscope. This increases the precision of the stereotaxic operation, and makes it possible to examine the probes for air bubbles and leaks. Microdialysis samples may be collected by changing collection vials by hand, or by using a microfraction collector which collects samples smaller than falling droplets by making the outlet tube from the probe touch the bottom of the sample vial. This is necessary because the sample volume is very seldom a n integer multiple of a droplet. The vials should be removed often, sealed and refrigerated to prevent evaporation and breakdown, or else a microfraction collector that collects under septa in closed refrigerated vials should be used. When
sampling for catecholamine analysis, a few microlitres of 1 M perchloric acid may be added to the vials prior to sampling in order to minimize decomposition of the samples. One way of avoiding both fraction collectors and autosamplers is to use an on-line injector, which is a common HPLC valve where the outlet of the probe is connected to the loop of the valve. The sample in the loop is then loaded into the chromatograph at regular intervals [ 3 5 ] . The advantage is that samples are analysed on line with the experiment. The flow rate through the microdialysis probe can be kept low, which increases the concentration and simplifies detection. The disadvantages are that the sampling time is dependent on the analysis time of the chromatographic method, and that samples cannot be divided and injected on to different chromatographic systems in order to increase the number of analytes. Microdialysis on conscious animals requires different instrumentation. It is important that the animal does not suffer from the trauma of probe implantation during the microdialysis experiment.
PRINCIPLES OF MICRODIALYSIS
This is traditionally avoided by prior implantation of a guide cannula and painless insertion of the probe once the animal has recovered. One can then perform a n ‘acute experiment’ in the conscious animal, or wait an extra 2 4 4 8 h if necessary to achieve the optimum conditions. However, an alternative is to implant the probe directly, without a guide cannula, and then wait for the required time. The probe can be sterilized prior to implantations by placing it in 70% alcohol. This procedure also dissolves the glycerol which is present in most membranes and improves diffusion. The limits of microdialysis are usually determined by the sensitivity of the analytical technique, even though it is now possible to analyse almost every known small neurotransmitter, metabolite and electrolyte by HPLC using electrochemical, fluorescence, UV and conductance detectors, sometimes in combination with enzyme reactors. However, there are still problems regarding the sensitivity of RIA for the detection of many peptides. One way to compensate for inadequate sensitivity is to increase the recovery by selecting a probe with the longest possible membrane and using the maximum acceptable sampling time. If the analytical technique is not sensitive to the volume of the sample, e.g. RIA, then the perfusion flow should be increased in order to increase the absolute recovery. However, if the analytical technique requires small sample volumes, e.g. microbore HPLC. the flow should be decreased in order to increase the relative recovery, i.e. the concentration of the sample. A second way to compensate for inadequate sensitivity is to increase the concentration of the substance to be analysed by pharmacological means. For all practical purposes it is still necessary to include an acetylcholine esterase inhibitor in the perfusate in order to recover adequate levels of acetylcholine, and most investigators include a reuptake blocker in the perfusate when studying serotonin. It would appear to be better to accept this compromise than not to study these transmitters at all. The important point is to conduct control experiments in such a manner that they reveal the action of the pharmacological treatment per se. A third way to compensate for inadequate sensitivity is to prelabel neurones by perfusion with isotopes of transmitters and their precursors. The sensitivity of the detection is not usually a problem, and the technique is well known from push-pull and slice experiments.
371
Future applications of microdialysis More than 600 articles have now been published on microdialysis experiments. Microdialysis has found its most important application in the neurosciences, because it developed out of a tradition of neuroscience techniques, and in response to a strong interest in linking the dramatic development of neuroanatomy, pharmacology, and physiological psychology to functional studies at the synaptic level. Its success is due to the parallel development of analytical chemistry. Drugs in combination with microdialysis are major tools. I am convinced that we will see microdialysis used as an important tool in the development of new drugs. The technique is capable of revealing the individual profile of the synaptic effects of a drug, and is becoming a promising tool in pharmacokinetic and drug distribution studies on animals and man. The use of microdialysis in behavioural studies is very promising indeed. Several studies have shown a correlation between behaviour and changes in transmitter levels in the dialysate [3640], but there have also been reports of a lack of correlation between such levels and behaviour [41]. Microdialysis has great potential for studies in peripheral organs including blood. Several studies have been published in which microdialysis was used in adipose tissue [9, 34,421 adrenal [43], blood [44], eye [45], heart [46], liver [47], muscle [12, 481, ovary [49] and uterus [SO]. In general, microdialysis is simple to perform in these organs because of their large size and homogenous parenchyma, making it possible to use longer membranes to increase recovery. The application in the uterus is of interest as it represents the first application of microdialysis in a body cavity, dialysing the epithelial secretion. Microdialysis can bridge the gap between animal models and man due to its simplicity and limited invasiveness. Studies on ischaemia after middle cerebral artery occlusion [S 11 have been compared with studies in man during ischaemia that developed in brain tissue during resection [lo], a model for concussive brain injury in rats [52] has been compared with intracerebral microdialysis in patients after skull trauma (Hillered et al., to be published), and rat studies on adipose tissue metabolism [9] have been extended to subcutaneous microdialysis in man ~421. Finally, I believe that microdialysis will develop into a routine sampling technique for clinical use. Its
372
U. U N G E R S T E D T
advantages are its limited invasiveness, the ability to sample continuously over long time periods without drawing blood, and the ‘new frontier’ whereby local organ chemistry is used for diagnostic and therapeutic purposes.
17
18
References 1 Gaddum JH. Push-pull cannulae. J Physiol 1961;155: 1-2. 2 Bito L. Davson H. Levin EM. Murray M. Snider N. The concentration of free amino acids and other electrolytes in cerebrospinal fluid, in vivo dialysate of brain and blood plasma of the dog. J Neurochem 1966;13: 1057-67. 3 Delgado JMR. DeFeudis FV. Roth RH. Ryugo DK. Mitruka BM. Dialytrode for long term intracerebral perfusion in awake monkeys. Arch lnt Pharmacodyn Ther 1972;198: 9-21. 4 Ungerstedt U. Pycock C. Functional correlates of dopamine neurotransmission. Bull Schweiz Akad Med Wiss 1974;1278: 1-1 3. 5 Ungerstedt U. Herrera-Marschitz M. Jungnelius U. Stdhle L. Tossman U. Zetterstrom T. Dopamine synaptic mechanisms reflected in studies combining behavioural recordings and brain dialysis. In: Kotisaka M. Shomori T. Tsukada Y. Woodruff GM. eds. Advances in Dopamine Research. New York: Pergamon Press, 1982:219-31. 6 Zeterstrom T. Sharp T. Marsden CA, Ungerstedt U. In vivo measurement ofdopamine and its metabolites by intracerebral dialysis : changes after d-amphetamine. / Neurochem 1984; 41 : 1769-73. 7 Ungerstedt U. Measurement of neurotransmitter release by intracranial dialysis. In : Marsden CA. ed. Measurement of Neurotransmitter Release in Vivo. Chichester: John Wiley & Son Ltd. 1984;81-105. 8 Tossman U, Ungerstedt U. Microdialysis in the study of extracellular levels of amino acids in the rat brain. Acta Physiol Scand 1986;128: 9-14. 9 Arner P. Bolinder J. Eliasson A. Lundin A, Ungerstedt U. Microdialysis of adipose tissue and blood for in vivo lipolysis studies. Am J Physiol 1988;255: 737-42. 10 Hillered L. Persson L. Ponten U. Ungerstedt U. Neurometabolic monitoring of the ischemic human brain using microdialysis. Acta Neurochir (Wien) 1990;102: 91-7. 1 1 Meyerson BA. Linderoth B. Karlsson H. Ungerstedt U. Microdialysis in the human brain :extracellular measurements in the thalamus of parkinsonian patients. Lije Sci 1990:46: 301-8. 12 Henriksson H. Fuchi T. Oshida Y. Ungerstedt U. Microdialysis of skeletal muscle for in vivo studies of cellular glucose metabolism. / Neurol Sci 1990;98: 250. 13 Benveniste H. Drejer J, Schousboe A, Diemer NH. Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem 1984: 43 : 1369-74. 14 Benveniste H. Brain microdialysis. J Neurochem 1989 ; 52 : 1667-79. 15 Amberg G. Lindefors N. Intracerebral microdialysis. 11. Mathematical studies of diffusion kinetics. J Pharmacol Methods 1989:22: 157-83. 16 Benveniste H, Drejer J. Schousboe A, Diemer NH. Regional
19
20
21
22
23
24
25
26
27
28
29
30
31
cerebral glucose phosphorylation and blood flow after insertion of a microdialysis fiber through the dorsal hippocampus in the rat. J Neurochem 1987:49: 729-34. Drew KL. O’Connor WT. Kehr 1. Ungerstedt U. Characterixation of gamma-aminobutyric acid and dopamine overflow following acute implantation of a microdialysis probe. Life Sci 1989;45: 1307-17. Osborne PG. O’Connor WT. Drew KL. Ungerstedt U. An in vivo microdialysis characterization of extracellular dopamine and GABA in dorsolateral striatum of awake freely moving and halothane anaesthetized rats. / Neurosci Methods 1990;34: 99-1 05. Westerink BHC. de Vries JB. Characterization of in vivo dopamine release as determined by brain microdialysis after acute and subchronic implantations : methodological aspects. J Neurochem 1988: 51 : 683-7. Imperato A, Di Chiara G. Trans-striatal dialysis coupled to reverse phase high performance liquid chromatography with electrochemical detection: a new method for the study of the in vivo release of endogenous dopamine and metabolites. / Neurosci 1984;4: 966-77. Westerink BHC. Hofsteede HM. Ilamsma G. de Vries JB, The significance of extracellular calcium for the release of dopamine. acetylcholine and amino acids in conscious rats, evaluated by brain microdialysis. Naunyn Schmiedebergs Arch Pharmacol 1988;337: 373-8. Moghaddam B. Bunney BS. Differential effect of cocaine on extracellular dopamine levels in rat medial prefrontal cortex and nucleus accumbens : comparison to amphetamine. Synapse 1989:4: 156-61. Lindefors N. Amberg G. Ungerstedt U. lntracerebral microdialysis. 1. Experimental studies of diffusion kinetics. / Pharmacol Methods 1989;22: 141-56. Bungay PM. Morrison PF. Dedrick RL. Steady-state theory for quantitative microdialysis of solutes and water in vivo and in vitro. Life Sci 1990;46: 105-19. Westerink BHC. De Vries IB. On the origin - of extracellular GABA collected by brain microdialysis and assayed by a simplified on-line method. Naunyn Schmiedbergs Arch Pharmacol 1989;339: 603-7. Kalen P. Strecker RE, Kosengren E. BiGrklund A. Endogenous release of neuronal serotonin and 5-hydroxyindoleacetic acid in the caudate-putamen of the rat as revealed by intracerebral dialysis coupled to high-performance liquid chromatography with fluorimetric detection. / Neurochem 1988;51 : 1422-35. Kalen P. Kakaia M. Lindvall 0. Bjorklund A. Basic characteristics of noradrenaline release in the hippocampus of intact and 6-hydroxydopamine-lesioned rats as studied by in vivo microdialysis. Brain Kes 1988;474: 374-9. Zetterstrom T,Ungerstedt U. Effects of apomorphine on the in vivo release of dopamine and its metabolites, studied by brain analysis. Eur J Pharmacol 1984:97: 29-36. Damsma G. Westerink BHC. De Vries JB, Van Den Berg CJ. Horn AS. Measurement of acetylcholine release in freely moving rats by means of automated intracerebral dialysis. / Neurochem 1987:48: 1523-8. Marien MR, Richard JW. Drug effects o n the release of endogenous acetylcholine in vivo: measurement by intracerebral dialysis and gas chromatography mass spectrometry. J Neurochem 1990;54: 2016-23. Kehr J , Ungerstedt U. Fast HPLC estimation of y-aminobutyric acid in microdialysis perfusates: effect of nipecotic and 3mercaptopropionic acids. / Neurochem 1988 ; 51 : 1308-1 0. I
P R I N C I P L E S OF M I C R O D I A L Y S I S
32 Reid M. O'Connor WT. Herrera-Marschitz M. Ungerstedt U. The effects of intranigral GABA and dynorphin A injections o n striatal dopamine and GABA release : evidence that dopamine provides inhibitory regulation of striatal GABA neurons via D, receptors. Brain Res 1 9 9 0 : 519: 255-60. 33 Jacobson I. Sandberg M. Hamberger A. Mass transfer in brain dialysis devices-a new method for the estimation of extracellular amino acid concentration. J Neurosci Methods 1 9 8 5 ; 1 5 : 263-8. 34 Liinnroth P. Jansson P-A. Smith U. A microdialysis method allowing characterization of intercellular water space in humans. Am J Physiol 1 9 8 7 : 253: 228-31. 35 Church WH. Justice JB, Neil1 DB. Detecting behavioral relevant changes in extracellular dopamine with microdialysis. Brain Res 1 9 8 7 ; 4 1 2 : 397-9. 36 Abercrombie ED, Keller RW. Zigmond MJ. Characterization of hippocampal norepinephrine release as pharmacological and behavioral studies. Neuroscience 1988 ; 2 7 : 897-904. 37 Click SD. Carlson JN. Baird JL. Maisonneuve IM. Bullock AE. Basal and amphetamine-induced asymmetries in striatal dopamine release and metabolism : bilateral in vivo microdialysis in normal rats. Bruin Res 1 9 8 8 : 4 7 3 : 1 6 1 4 . 38 Hernandez L. Hoebel BG. Food reward and cocaine increase extracellular dopamine in the nucleus accumbens as measured by microdialysis. Lije Sci 1988 : 4 2 : 1 705-1 2. 39 Kendrick KM. Keverne EB, Chapman C, Baldwin BA. Intracranial dialysis measurement of oxytocin, monoamine and uric acid release from the olfactory bulb and substantia nigra of sheep during parturition, suckling, separation from lambs and eating. Brain Res 1 9 8 8 ; 4 3 9 : 1-10, 40 Radhakishun FS. Van Ree JM, Westerink BHC. Scheduled eating increases dopamine release in the nucleus accumbens of food deprived rats as assessed with on-line brain dialysis. Neurosci Lett 1 9 8 8 ; 85: 351-6. 41 StAhIe L. Ungerstedt U. Reduction of extracellular dopamine levels can be dissociated from suppression of exploratory behavior in rats. Acta Physiol Scand 1 9 8 7 ; 130: 533-4. 4 2 Arner P, Kriegholm E. Engfeldt P. In situ studies of catecholamine-induced lipolysis in human adipose using microdialysis. J Pharmacol Erp "her 1 9 9 0 : 254: 284-8. 4 3 Jarry H. Duher EM, Wuttke W. Adrenal release of
44
45
46
47
48
49
50
51
52
373
catecholamine and metenkephalin before and after stress as measured by a novel in vivo dialysis method on the rat. Neurosci Lett 1 9 8 5 ; 6 0 : 272. Speciale C. Ungerstedt U. Schwarcz R. Effect of kynurenine loading on quinolinic acid production in the rat: studies in vitro and in vivo. Lije Sci 1 9 8 8 : 4 3 : 777-86. Ben-Nun J, Joyce DA. Cooper RL. Cringle SJ, Constable J . Pharmacokinetics of intravitreal injection. Invest Ophthalmol Vis Sci 1989: 30: 1055-61. Hamberger A. Microdialysis in clinical diagnosis: amino acid patterns in the temporal cortex and the myocardium. Curr Separations 1 9 8 9 : 9: 119. Scott DO, Sorensen LR. Lunte CE. In vivo microdialysis sampling coupled to liquid chromatography for the study of acetaminophen metabolism. J Chromatogr 1 9 9 0 ; 506: 461-9. k h m a n n A. Effects of microdialysis-perfusion with anisosmotic media on extracellular amino acids in the rat hippocampus and skeletal muscle. 1 Neurochem 1 9 8 9 : 51 : 52 5-3 5. Jarry H. Einsparrier A, Kannpiesser I, et al. Release and effects of oxytocin on estradiol and progesterone secretion in porcine corpora lutea as measured by an in vivo microdialysis system. Endocrinology 1 9 9 0 ; 1 2 6 : 2350-8. Nordenvall M, Ulmsten U, Ungerstedt U. Influence of progesterone on the sodium and potassium concentrations of rat uterine fluid investigated by microdialysis. Gynecol Obstet Invest 1 9 8 9 ; 2 8 : 73-7. Hillered L. Hallstrom A. Segersvlrd S. Persson L. Ungerstedt U. Dynamics of extracellular metabolites in the striatum after middle cerebral artery occlusion on the rat monitored by intracerebral microdialysis. I Cereb Blood Flow Metab 1 9 8 9 : 9: 607-16. Nilsson P. Hillered L. Ponten U. Ungerstedt U. Changes in cortical extracellular levels of energy-related metabolites and amino acids following concussive brain injury in rats. I Cereb Blood Flow Metab 1 9 9 0 ; 10: 631-7.
Received 15 April 1991, accepted 7 May 1991. Correspondence: Prof. U. Ungerstedt. Department of Pharmacology, Karolinska Institutet. 104 04 Stockholm, Sweden.
Microdialysis-principles animals and man
and applications for studies in
U. UNGERSTEDT From the Department o/ Pharmacology. Karolinska institute. Stockholm. Sweden
Abstract. Ungerstedt U (Department of Pharmacology, Karolinska Institute, Stockholm. Sweden), Microdialysis-principles and applications for studies in animals and man. Journal of lnternal Medicine 1991 : 230: 365-373
Microdialysis is a technique for sampling the chemistry of the individual tissues and organs of the body, and is applicable to both animal and human studies. The basic principle is to mimic the function of a capillary blood vessel by perfusing a thin dialysis tube implanted into the tissue with a physiological liquid. The perfusate is analysed chemically and reflects the composition of the extracellular fluid with time due to the diffusion of substances back and forth over the membrane. Microdialysis is thus a technique whereby substances may be both recovered from and supplied to a tissue. The most important features of microdialysis are as follows: it samples the extracellular fluid, which is the origin of all blood chemistry; it samples continuously for hours or days without withdrawing blood ; and it purifies the sample and simplifies chemical analysis by excluding large molecules from the perfusate. However, the latter feature renders the technique unsuitable for sampling large molecules such as proteins. The technique has been extensively used in the neurosciences to monitor neurotransmitter release, and is now finding application in monitoring of the chemistry of peripheral tissues in both animal and human studies. Keywords : extracellular fluid, microdialysis.
Introduction In medical practice, most sampling from the body is achieved by drawing blood. It is relatively straightforward, acceptable, and the reference ranges are well known. However, blood merely reflects the chemical composition of the extracellular fluid of the individual organs. In many instances it would have been more relevant and informative to sample this fluid directly. Due to the development of a new technique, microdialysis, this is now possible. Most experimental attempts to sample from the extracellular environment have been performed on the brain. Gaddum [l] devised a ‘push-pull’ technique whereby liquid was pumped (pushed) directly into the brain tissue through one cannula, while it was simultaneously withdrawn (pulled) from the tissue at the same speed by an adjacent cannula. However, the push-pull technique suffers from the drawback that the tissue is disturbed by the streaming
liquid-it is necessary to balance the push and pull flows exactly in order to avoid sucking out tissue or injecting liquid. Bit0 et al. [2] were the first to implant ‘dialysis sacs ’ containing 6% dextran in saline into the subcutaneous tissue of the neck and the parenchyma of the cerebral hemispheres of dogs. Several weeks later they removed the sacs surgically and analysed their amino acid content. These studies introduced the concept of a ‘compartment’ surrounded by a dialysis membrane which equilibrates with the extracellular environment. However, the intention was not to monitor changes over time, but to avoid physiological fluctuations and to obtain an average concentration of the substances of interest. Delgado et al. [ 3 ] developed a ‘dialytrode’ consisting of two stainless steel tubes forming a pushpull cannula ending in a small permeable bag made of dialysis membrane. The dialytrode was tested in vivo in monkey brain tissue. 365
366
U. UNGERSTEDT
In our own laboratory we implanted ‘hollow fibres’ into the brain in order to mimic the function of blood vessels. When dopamine neurones were labelled by perfusion with 3H-dopamine we could monitor baseline and amphetamine-stimulated release in anaesthetized and freely moving animals in relation to behavioural activation [4]. The simple hollow fibre ‘ microdialysis probe ’ has subsequently been extensively improved, so that it is now possible to implant a ‘needle probe’ (Fig. 1) with a hollow fibre membrane at its tip [5-81 into brain and other tissues in much the same manner as a cannula or an electrode. We have applied this probe to studies on subcutaneous [9], brain [lo, 111 and muscle tissue [12] in man.
Features of microdialysis The basic principle of microdialysis is to mimic the passive function of a capillary blood vessel by perfusing a thin dialysis tube implanted into the tissue. The perfusate is analysed chemically and reflects the composition of the extracellular fluid over time due to the diffusion of substances back and forth over the membrane. Microdialysis is thus a technique whereby substances may be both recovered from and administered to a tissue. The most important features of microdialysis are as follows. (i) It samples the extracellular fluid as distinct from the whole tissue collected by biopsies, punches or dissection. (ii) It can be performed locally in almost every organ and tissue of the body (including blood). (iii) It makes it possible to sample continuously for hours or days. (iv) It can be used to recover and/or introduce endogenous and exogenous substances in the tissue. For example, it is possible to administer a drug systemically while using microdialysis to estimate the local drug concentration, the local biochemical effect and the resulting physiological response. (v) It collects a representative sample of all substances in the extracellular fluid (provided that they pass across the membrane), carries them out of the body and makes them accessible to conventional analytical techniques. This distinguishes the technique from in-vivo sensor methods, such as implantable biosensors. (vi) It causes minimal damage to the blood-brain barrier [8, 131. This makes it possible (a) to
study drug penetration into the brain by analysing the drug concentration in the perfusate and (b) to compare the effects of drugs on the brain when applied directly via the microdialysis probe or via the systemic route. (vii) It simplifies perfusions. The flow is unidirectional, and there is no need to balance the flow as in a push-pull cannula. (viii) It minimizes damage to the tissue, as there is no direct contact between the fluid flowing outside the membrane and the cells of the tissue. (ix) The membrane prevents large molecules from diffusing into the perfusate. This protects certain molecules from degradation by tissue enzymes once they have been recovered from the tissue. (x)The process of microdialysis purifies the sample, so that it is possible to introduce it directly into the analyticai instrument, for example a n HPLC system. (xi) It introduces a sterility barrier between the perfusion liquid and the tissue. (xii) The size of the perfused area can be regulated by varying the length of the membrane. (xiii) Microdialysis can be performed in blood or in body cavities such as the uterus, the peritoneal cavity and the mouth. This is impossible with an open push-pull system. (xiv) The performance of a microdialysis probe and its membrane can be studied in vitro. Different probes can be compared with regard to the recovery of a particular substance, or a single probe can be studied for its ability to recover substances of different molecular weight or charge. (xv) Clinically, microdialysis allows repeated sampling for hours or days after a single penetration of the tissue, without the need to withdraw or handle any blood.
Principles of microdialysis Microdialysis is a very simple technique. A tubular dialysis membrane is introduced into the tissue or placed in contact with a moist surface, for example a mucous membrane. The tube is perfused with a liquid that equilibrates with the fluid outside the tube by diffusion in both directions. The degree of equilibration is subject to known laws of physical chemistry. The complexity of the technique is due to the interaction between the dialysis tube, the perfusion
PRINCIPLES OF MICRODIALYSIS
0
.O 0 0 '
0
0
0
.
0
0 0
Fig. 1. Microdialysis probe. The perfusion liquid enters through the right inlet cannula and passes through an inner cannula all the way to the tip of the probe (see enlargement of the tip) where it flows upwards between the inner cannula and the membrane, where dialysis takes place. The liquid leaves the probe through the left outlet cannula.
liquid and the living tissue. In order to perform microdialysis and to interpret the results correctly, it is important to envisage the many events that take place during this interaction. There is a rapid fall in the concentration of most substances in the perfusate during the first phase of microdialysis. which is probably due not only to an initial lesion of the tissue which causes excessive release of substances from cellular storage compartments, but also to the establishment of a new steady-state level of most extracellular substances because of drainage through the probe [14. 151. In our experience the degree of initial damage determines how rapidly baseline levels are reached, and care should be taken to introduce the probe slowly into the tissue. During the initial period of disturbed tissue function glucose metabolism is increased, blood flow decreased, and transmitter release is disrupted. This period lasts from minutes to hours, depending on the
367
tissue, the functions of which are measured and the substances analysed [16-181. Baseline levels are reached more rapidly in larger organs where the relative size of the probe is smaller. In the human brain this often occurs within 10-20 min [Ill. A related problem is the time period over which substances can be reliably recovered once they have reached 'baseline ' levels. This varies between organs, substances and investigators [18. 191. Such variation is not surprising in view of the biological differences between tissues with regard to synthesis, storage, metabolism, density of innervation, delicacy of nerve terminals, regional difference in blood flow, etc. There are also obvious differences between laboratories and the types of probe employed, use of guide cannula, maintenance of sterility, implantation procedures, composition of the perfusion medium, biocompatibility of membrane, species, etc. In clinical experiments we have been able to maintain constant levels of various metabolites in the subcutaneous tissue for 6-7 d (unpublished data). The microdialysis probe communicates with the extracellular fluid by diffusion along a logarithmic concentration gradient towards and away from the probe. The direction of this gradient is dependent on the composition of the perfusion liquid, and the latter dominates the immediate environment of the probe. If calcium is excluded from the perfusion liquid, the extracellular fluid surrounding the probe is depleted of calcium, impairing synaptic transmission [20, 2 11. Conversely, inclusion of a substance in the perfusate will make it spread from the probe along a concentration gradient the direction of which is opposite to that of calcium depletion. The composition of the perfusate should be as close as possible to normal physiological levels of the most essential compounds in the extracellular environment [22]. The extent of this similarity must be determined for individual cases, as the aim of a microdialysis experiment is to remove or administer substances by exploiting the differeqres between the perfusion liquid and the extracellular fluid. The recovery of substances from the extracellular fluid is dependent on (i) the length of the dialysis membrane, (ii) the flow of the perfusion liquid, (iii) the speed of diffusion of the substance through the extracellular fluid and (iv) the properties of the membrane [15, 14, 23, 241. The speed of diffusion is a property related to diffusion through the parenchyma as well as to active elimination from the tissue by uptake into cells and blood capillaries. The
368
U. UNGERSTEDT
recovery is the same in both directions across the membrane [15]. For small molecules, such as the monoamine transmitters, the factor that limits recovery is usually the speed of diffusion through the extracellular fluid, and not diffusion through the membrane [ 151. In the tortuous brain parenchyma the speed of diffusion may be around 50% of that in water, while the corresponding figure for the membrane may be as high as 80%. Although it is tempting to think that the membrane represents an important barrier, it is actually the ability of a substance to diffuse through the extracellular space that determines both the amount of substance that can be recovered and the speed with which a change can be detected. It is convenient to distinguish between relative recovery (concentration recovery) and absolute recovery (mass recovery) [7]. Relative recovery is the concentration of a particular substance in the perfusate when it leaves the probe, expressed as a percentage of the concentration in the surrounding medium. Absolute recovery is the total amount of substance recovered during a defined time period, expressed in mol I-'. When a microdialysis probe is tested in vitro, relative recovery decreases and absolute recovery increases as the perfusion flow is increased. Both values reach a plateau when the diffusion speed through the surrounding medium reaches its maximum. In vivo the relative recovery is constant as long as the perfusion conditions remain the same. However, the absolute recovery of a substance varies with its production/release in the tissue. The usual reason for performing microdialysis is to monitor this change in the extracellular levels of endogenous or exogenous substances.
Neurotransmitter release A central question is whether recovered transmitters reflect ' true ' synaptic release or a more unspecific overflow from synaptic and non-synaptic sources. Blocking of the dopamine [25]. 5-HT [26] or noradrenaline [2 71 reuptake mechanisms increases the level of the transmitters in the perfusate, and stimulation of dopamine autoreceptors [2 81 gives the expected decrease in dopamine. Perfusion with TTX, in order to block sodium channels, lowers acetylcholine [29], dopamine [191 and noradrenaline [2 71 levels. The omission of calcium lowers dopamine [20], 5-HT [26] and acetylcholine [30] levels.
Although these data strongly suggest that the microdialysis probe recovers an overflow from the synaptic release, they do not prove that this overflow is quantitatively related to synaptic release. Studies of GABA are more complicated. Uptake block increases GABA in the perfusate [31], while the effect of low calcium and TTX varies according to the smoothness of the implantation procedure, the time period between implantation of the probe and the experiment, and whether the animal is anaesthetized or conscious. Drew et al. [17] and Westerink and De Vries [25] did not find any effects of TTX in acutely implanted, anaesthetized animals, while Osborne et al. [18] were able to demonstrate decreased GABA after both TTX and omission of calcium in conscious animals in which the microdialysis probes had been implanted for 1 4 d. The correct interpretation of the GABA experiments is probably that GABA in the perfusate does not originate solely from synaptic sources. However, the fact that GABA in the perfusate is increased by dopamine D1 receptor agonists and lowered by D2 agonists [32] seems to provide positive evidence for release under synaptic control.
Finding the ' true ' extracellular concentration Another fundamental problem is the relationship between the concentration in the perfusate and that in the extracellular fluid. Several attempts have been made to determine this concentration by extrapolation from in-vitro recovery values, but there is now general agreement that this is not reliable because of the difference in diffusion coefficients between water and tissue. The best estimates of extracellular concentrations are obtained from in-vivo experiments. They can be calculated by varying the perfusion flow during a n in-vivo experiment, measuring the change in the substance of interest coming out of the probe, and then extrapolating to zero flow [33], or by perfusing with varying concentrations of the substance and then calculating the equilibrium concentration, i.e. the concentration at which the substance in the perfusate does not change during the perfusion because it has the same concentration inside the probe as in the extracellular fluid [34]. Both these invivo methods require that the concentration in the extracellular fluid remains constant throughout the experiment.
PRINCIPLES O F MICRODJALYSIS
A third method is to use a reference substance in the perfusate, or to perfuse with it before the systemic injection of an exogenous compound (StBhle. to be published). The method is based on the fact that recovery over the membrane is the same in both directions. Given that the diffusion properties of the reference substance in the perfusate and the substance to be recovered from the brain are the same, the percentage loss of the reference substance from the perfusate will be the same as the percentage recovery of the substance from the brain. This recovery value is then used to calculate the concentration in the extracellular fluid. Thus in pharmacokinetic experiments the true extracellular concentration of a drug can be determined by equilibrium dialysis ' during steady-state conditions, or by including a reference substance in the perfusate. This substance should have the same diffusion characteristics as the systemically administered drug, ideally appear as a peak in the same chromatographic system, and be pharmacologically inactive at the concentration used.
The microdialysis probe The reason for making a microdialysis probe is to provide a convenient means of introducing a dialysis tube into the tissue. There are basically three types of probe : the dialysis tube per se, the loop probe and the concentric probe. The advantage of using the dialysis tube per se is that it is simple to make such a probe. The drawback is that it is necessary to make holes for both entry and exit. The loop probe consists of two parallel metal tubes, connected by a dialysis tube forming a loop [6]. In order to stretch the loop during implantation into the tissue, a stylus is placed inside or alongside one of the metal tubes, reaching all the way to the distal end of the loop. The loop probe is easy to make and position in a tissue. However, the two parallel tubes increase its size and thus enhance the damage to the tissue. In the concentric probe [8] a piece of dialysis tubing is sealed by glue at one end forming the tip of the probe (Fig. 1).The other end is usually glued into a tube which is the shaft of the probe. A thin inner cannula extends through the shaft and dialysis tube all the way to the tip. The material may be metal, fused silica or plastic. The perfusing liquid usually enters the proximal end of the inner cannula and flows distally all the way to its end, where it changes
369
direction and returns in the space between the inner cannula extends and the membrane where microdialysis takes place. The proximal end is designed in such a way that the inlet and outlet are separated. The concentric probe is the most difficult to construct. It can be made very thin, and the length of the tip can be varied from mm to cm in length. The perfusion liquid is carried to and from the probe by thin tubes, the importance of which is often neglected. They usually represent a substantial inner volume which delays the introduction of substances into the probe, as well as collection of the samples. When using a low perfusion flow rate it may be many minutes from the time the drug is included in the perfusate until it reaches the probe in the tissue, and still longer before it reaches the collecting vial. The material of the tubing should be as inert as possible in order to avoid adsorption of substances to its walls. The perfusion flow rate should be low (0.1-5 pl min-') in order to remove as little as possible of the extracellular content and thus minimize the interference with normal physiology. This demand must be balanced against the difficulty encountered in handling very small amounts of fluid and the need to obtain enough material for the analytical technique. In large homogenous organs such as the liver or adipose tissue, it is possible to compensate for a low flow by increasing the length of the membrane. Under ideal conditions, e.g. during microdialysis in the blood, the outer environment remains constant due to the streaming blood. With a 20-mm membrane in a vein of a pig the recovery of lactate is virtually 100% at a flow rate of 2-3 1 1 min-'.
Microdialysis instruments The technical difficulties of microdialysis relate to the handling of low flow rates and small volumes. However, most syringe pumps on the market can produce the required flow rates. A few instruments have been developed specifically for microdialysis, such as the CMA 100 series, consisting of a syringe pump which powers and controls a syringe selector, a microfraction collector or an injector for on-line injections into a chromatograph. The CMA 200 series includes a refrigerated fraction collector in which the sample cassettes can be transferred directly to a refrigerated autosampler and the samples injected into one or two different chromatographs.
370
U. UNGEKSTEDT
Fig. 2. Schematic drawing of the instruments used during a microdialysis experiment on a rat. A = computer controlling the experiment, B = perfusion pump with three syringes containing different perfusion fluids, C = syringe selector which selects one of the syringes of the pump (B) under computer control (A), D = anaesthetixed rat in a stereotaxic instrument. The microdialysis probe is held in the instrument and placed in the brain, E = microfraction collector, which collects fractions under the control of the computer (A) or the pump (B). F = second pump used to inject a drug systemically under the control of the computer (A), G = temperature controller to maintain the anaesthetized animal at the correct temperature.
The use of a syringe selector is important if the introduction of air bubbles into the perfusate, which often occurs if the tubing is transferred manually between syringes when changing the perfusion liquid (Fig. 2). is to be avoided. A stereotaxic instrument is necessary for microdialysis studies on the brain. It is a great advantage if the stereotaxic implantations can be performed under a stereo microscope. This increases the precision of the stereotaxic operation, and makes it possible to examine the probes for air bubbles and leaks. Microdialysis samples may be collected by changing collection vials by hand, or by using a microfraction collector which collects samples smaller than falling droplets by making the outlet tube from the probe touch the bottom of the sample vial. This is necessary because the sample volume is very seldom a n integer multiple of a droplet. The vials should be removed often, sealed and refrigerated to prevent evaporation and breakdown, or else a microfraction collector that collects under septa in closed refrigerated vials should be used. When
sampling for catecholamine analysis, a few microlitres of 1 M perchloric acid may be added to the vials prior to sampling in order to minimize decomposition of the samples. One way of avoiding both fraction collectors and autosamplers is to use an on-line injector, which is a common HPLC valve where the outlet of the probe is connected to the loop of the valve. The sample in the loop is then loaded into the chromatograph at regular intervals [ 3 5 ] . The advantage is that samples are analysed on line with the experiment. The flow rate through the microdialysis probe can be kept low, which increases the concentration and simplifies detection. The disadvantages are that the sampling time is dependent on the analysis time of the chromatographic method, and that samples cannot be divided and injected on to different chromatographic systems in order to increase the number of analytes. Microdialysis on conscious animals requires different instrumentation. It is important that the animal does not suffer from the trauma of probe implantation during the microdialysis experiment.
PRINCIPLES OF MICRODIALYSIS
This is traditionally avoided by prior implantation of a guide cannula and painless insertion of the probe once the animal has recovered. One can then perform a n ‘acute experiment’ in the conscious animal, or wait an extra 2 4 4 8 h if necessary to achieve the optimum conditions. However, an alternative is to implant the probe directly, without a guide cannula, and then wait for the required time. The probe can be sterilized prior to implantations by placing it in 70% alcohol. This procedure also dissolves the glycerol which is present in most membranes and improves diffusion. The limits of microdialysis are usually determined by the sensitivity of the analytical technique, even though it is now possible to analyse almost every known small neurotransmitter, metabolite and electrolyte by HPLC using electrochemical, fluorescence, UV and conductance detectors, sometimes in combination with enzyme reactors. However, there are still problems regarding the sensitivity of RIA for the detection of many peptides. One way to compensate for inadequate sensitivity is to increase the recovery by selecting a probe with the longest possible membrane and using the maximum acceptable sampling time. If the analytical technique is not sensitive to the volume of the sample, e.g. RIA, then the perfusion flow should be increased in order to increase the absolute recovery. However, if the analytical technique requires small sample volumes, e.g. microbore HPLC. the flow should be decreased in order to increase the relative recovery, i.e. the concentration of the sample. A second way to compensate for inadequate sensitivity is to increase the concentration of the substance to be analysed by pharmacological means. For all practical purposes it is still necessary to include an acetylcholine esterase inhibitor in the perfusate in order to recover adequate levels of acetylcholine, and most investigators include a reuptake blocker in the perfusate when studying serotonin. It would appear to be better to accept this compromise than not to study these transmitters at all. The important point is to conduct control experiments in such a manner that they reveal the action of the pharmacological treatment per se. A third way to compensate for inadequate sensitivity is to prelabel neurones by perfusion with isotopes of transmitters and their precursors. The sensitivity of the detection is not usually a problem, and the technique is well known from push-pull and slice experiments.
371
Future applications of microdialysis More than 600 articles have now been published on microdialysis experiments. Microdialysis has found its most important application in the neurosciences, because it developed out of a tradition of neuroscience techniques, and in response to a strong interest in linking the dramatic development of neuroanatomy, pharmacology, and physiological psychology to functional studies at the synaptic level. Its success is due to the parallel development of analytical chemistry. Drugs in combination with microdialysis are major tools. I am convinced that we will see microdialysis used as an important tool in the development of new drugs. The technique is capable of revealing the individual profile of the synaptic effects of a drug, and is becoming a promising tool in pharmacokinetic and drug distribution studies on animals and man. The use of microdialysis in behavioural studies is very promising indeed. Several studies have shown a correlation between behaviour and changes in transmitter levels in the dialysate [3640], but there have also been reports of a lack of correlation between such levels and behaviour [41]. Microdialysis has great potential for studies in peripheral organs including blood. Several studies have been published in which microdialysis was used in adipose tissue [9, 34,421 adrenal [43], blood [44], eye [45], heart [46], liver [47], muscle [12, 481, ovary [49] and uterus [SO]. In general, microdialysis is simple to perform in these organs because of their large size and homogenous parenchyma, making it possible to use longer membranes to increase recovery. The application in the uterus is of interest as it represents the first application of microdialysis in a body cavity, dialysing the epithelial secretion. Microdialysis can bridge the gap between animal models and man due to its simplicity and limited invasiveness. Studies on ischaemia after middle cerebral artery occlusion [S 11 have been compared with studies in man during ischaemia that developed in brain tissue during resection [lo], a model for concussive brain injury in rats [52] has been compared with intracerebral microdialysis in patients after skull trauma (Hillered et al., to be published), and rat studies on adipose tissue metabolism [9] have been extended to subcutaneous microdialysis in man ~421. Finally, I believe that microdialysis will develop into a routine sampling technique for clinical use. Its
372
U. U N G E R S T E D T
advantages are its limited invasiveness, the ability to sample continuously over long time periods without drawing blood, and the ‘new frontier’ whereby local organ chemistry is used for diagnostic and therapeutic purposes.
17
18
References 1 Gaddum JH. Push-pull cannulae. J Physiol 1961;155: 1-2. 2 Bito L. Davson H. Levin EM. Murray M. Snider N. The concentration of free amino acids and other electrolytes in cerebrospinal fluid, in vivo dialysate of brain and blood plasma of the dog. J Neurochem 1966;13: 1057-67. 3 Delgado JMR. DeFeudis FV. Roth RH. Ryugo DK. Mitruka BM. Dialytrode for long term intracerebral perfusion in awake monkeys. Arch lnt Pharmacodyn Ther 1972;198: 9-21. 4 Ungerstedt U. Pycock C. Functional correlates of dopamine neurotransmission. Bull Schweiz Akad Med Wiss 1974;1278: 1-1 3. 5 Ungerstedt U. Herrera-Marschitz M. Jungnelius U. Stdhle L. Tossman U. Zetterstrom T. Dopamine synaptic mechanisms reflected in studies combining behavioural recordings and brain dialysis. In: Kotisaka M. Shomori T. Tsukada Y. Woodruff GM. eds. Advances in Dopamine Research. New York: Pergamon Press, 1982:219-31. 6 Zeterstrom T. Sharp T. Marsden CA, Ungerstedt U. In vivo measurement ofdopamine and its metabolites by intracerebral dialysis : changes after d-amphetamine. / Neurochem 1984; 41 : 1769-73. 7 Ungerstedt U. Measurement of neurotransmitter release by intracranial dialysis. In : Marsden CA. ed. Measurement of Neurotransmitter Release in Vivo. Chichester: John Wiley & Son Ltd. 1984;81-105. 8 Tossman U, Ungerstedt U. Microdialysis in the study of extracellular levels of amino acids in the rat brain. Acta Physiol Scand 1986;128: 9-14. 9 Arner P. Bolinder J. Eliasson A. Lundin A, Ungerstedt U. Microdialysis of adipose tissue and blood for in vivo lipolysis studies. Am J Physiol 1988;255: 737-42. 10 Hillered L. Persson L. Ponten U. Ungerstedt U. Neurometabolic monitoring of the ischemic human brain using microdialysis. Acta Neurochir (Wien) 1990;102: 91-7. 1 1 Meyerson BA. Linderoth B. Karlsson H. Ungerstedt U. Microdialysis in the human brain :extracellular measurements in the thalamus of parkinsonian patients. Lije Sci 1990:46: 301-8. 12 Henriksson H. Fuchi T. Oshida Y. Ungerstedt U. Microdialysis of skeletal muscle for in vivo studies of cellular glucose metabolism. / Neurol Sci 1990;98: 250. 13 Benveniste H. Drejer J, Schousboe A, Diemer NH. Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem 1984: 43 : 1369-74. 14 Benveniste H. Brain microdialysis. J Neurochem 1989 ; 52 : 1667-79. 15 Amberg G. Lindefors N. Intracerebral microdialysis. 11. Mathematical studies of diffusion kinetics. J Pharmacol Methods 1989:22: 157-83. 16 Benveniste H, Drejer J. Schousboe A, Diemer NH. Regional
19
20
21
22
23
24
25
26
27
28
29
30
31
cerebral glucose phosphorylation and blood flow after insertion of a microdialysis fiber through the dorsal hippocampus in the rat. J Neurochem 1987:49: 729-34. Drew KL. O’Connor WT. Kehr 1. Ungerstedt U. Characterixation of gamma-aminobutyric acid and dopamine overflow following acute implantation of a microdialysis probe. Life Sci 1989;45: 1307-17. Osborne PG. O’Connor WT. Drew KL. Ungerstedt U. An in vivo microdialysis characterization of extracellular dopamine and GABA in dorsolateral striatum of awake freely moving and halothane anaesthetized rats. / Neurosci Methods 1990;34: 99-1 05. Westerink BHC. de Vries JB. Characterization of in vivo dopamine release as determined by brain microdialysis after acute and subchronic implantations : methodological aspects. J Neurochem 1988: 51 : 683-7. Imperato A, Di Chiara G. Trans-striatal dialysis coupled to reverse phase high performance liquid chromatography with electrochemical detection: a new method for the study of the in vivo release of endogenous dopamine and metabolites. / Neurosci 1984;4: 966-77. Westerink BHC. Hofsteede HM. Ilamsma G. de Vries JB, The significance of extracellular calcium for the release of dopamine. acetylcholine and amino acids in conscious rats, evaluated by brain microdialysis. Naunyn Schmiedebergs Arch Pharmacol 1988;337: 373-8. Moghaddam B. Bunney BS. Differential effect of cocaine on extracellular dopamine levels in rat medial prefrontal cortex and nucleus accumbens : comparison to amphetamine. Synapse 1989:4: 156-61. Lindefors N. Amberg G. Ungerstedt U. lntracerebral microdialysis. 1. Experimental studies of diffusion kinetics. / Pharmacol Methods 1989;22: 141-56. Bungay PM. Morrison PF. Dedrick RL. Steady-state theory for quantitative microdialysis of solutes and water in vivo and in vitro. Life Sci 1990;46: 105-19. Westerink BHC. De Vries IB. On the origin - of extracellular GABA collected by brain microdialysis and assayed by a simplified on-line method. Naunyn Schmiedbergs Arch Pharmacol 1989;339: 603-7. Kalen P. Strecker RE, Kosengren E. BiGrklund A. Endogenous release of neuronal serotonin and 5-hydroxyindoleacetic acid in the caudate-putamen of the rat as revealed by intracerebral dialysis coupled to high-performance liquid chromatography with fluorimetric detection. / Neurochem 1988;51 : 1422-35. Kalen P. Kakaia M. Lindvall 0. Bjorklund A. Basic characteristics of noradrenaline release in the hippocampus of intact and 6-hydroxydopamine-lesioned rats as studied by in vivo microdialysis. Brain Kes 1988;474: 374-9. Zetterstrom T,Ungerstedt U. Effects of apomorphine on the in vivo release of dopamine and its metabolites, studied by brain analysis. Eur J Pharmacol 1984:97: 29-36. Damsma G. Westerink BHC. De Vries JB, Van Den Berg CJ. Horn AS. Measurement of acetylcholine release in freely moving rats by means of automated intracerebral dialysis. / Neurochem 1987:48: 1523-8. Marien MR, Richard JW. Drug effects o n the release of endogenous acetylcholine in vivo: measurement by intracerebral dialysis and gas chromatography mass spectrometry. J Neurochem 1990;54: 2016-23. Kehr J , Ungerstedt U. Fast HPLC estimation of y-aminobutyric acid in microdialysis perfusates: effect of nipecotic and 3mercaptopropionic acids. / Neurochem 1988 ; 51 : 1308-1 0. I
P R I N C I P L E S OF M I C R O D I A L Y S I S
32 Reid M. O'Connor WT. Herrera-Marschitz M. Ungerstedt U. The effects of intranigral GABA and dynorphin A injections o n striatal dopamine and GABA release : evidence that dopamine provides inhibitory regulation of striatal GABA neurons via D, receptors. Brain Res 1 9 9 0 : 519: 255-60. 33 Jacobson I. Sandberg M. Hamberger A. Mass transfer in brain dialysis devices-a new method for the estimation of extracellular amino acid concentration. J Neurosci Methods 1 9 8 5 ; 1 5 : 263-8. 34 Liinnroth P. Jansson P-A. Smith U. A microdialysis method allowing characterization of intercellular water space in humans. Am J Physiol 1 9 8 7 : 253: 228-31. 35 Church WH. Justice JB, Neil1 DB. Detecting behavioral relevant changes in extracellular dopamine with microdialysis. Brain Res 1 9 8 7 ; 4 1 2 : 397-9. 36 Abercrombie ED, Keller RW. Zigmond MJ. Characterization of hippocampal norepinephrine release as pharmacological and behavioral studies. Neuroscience 1988 ; 2 7 : 897-904. 37 Click SD. Carlson JN. Baird JL. Maisonneuve IM. Bullock AE. Basal and amphetamine-induced asymmetries in striatal dopamine release and metabolism : bilateral in vivo microdialysis in normal rats. Bruin Res 1 9 8 8 : 4 7 3 : 1 6 1 4 . 38 Hernandez L. Hoebel BG. Food reward and cocaine increase extracellular dopamine in the nucleus accumbens as measured by microdialysis. Lije Sci 1988 : 4 2 : 1 705-1 2. 39 Kendrick KM. Keverne EB, Chapman C, Baldwin BA. Intracranial dialysis measurement of oxytocin, monoamine and uric acid release from the olfactory bulb and substantia nigra of sheep during parturition, suckling, separation from lambs and eating. Brain Res 1 9 8 8 ; 4 3 9 : 1-10, 40 Radhakishun FS. Van Ree JM, Westerink BHC. Scheduled eating increases dopamine release in the nucleus accumbens of food deprived rats as assessed with on-line brain dialysis. Neurosci Lett 1 9 8 8 ; 85: 351-6. 41 StAhIe L. Ungerstedt U. Reduction of extracellular dopamine levels can be dissociated from suppression of exploratory behavior in rats. Acta Physiol Scand 1 9 8 7 ; 130: 533-4. 4 2 Arner P, Kriegholm E. Engfeldt P. In situ studies of catecholamine-induced lipolysis in human adipose using microdialysis. J Pharmacol Erp "her 1 9 9 0 : 254: 284-8. 4 3 Jarry H. Duher EM, Wuttke W. Adrenal release of
44
45
46
47
48
49
50
51
52
373
catecholamine and metenkephalin before and after stress as measured by a novel in vivo dialysis method on the rat. Neurosci Lett 1 9 8 5 ; 6 0 : 272. Speciale C. Ungerstedt U. Schwarcz R. Effect of kynurenine loading on quinolinic acid production in the rat: studies in vitro and in vivo. Lije Sci 1 9 8 8 : 4 3 : 777-86. Ben-Nun J, Joyce DA. Cooper RL. Cringle SJ, Constable J . Pharmacokinetics of intravitreal injection. Invest Ophthalmol Vis Sci 1989: 30: 1055-61. Hamberger A. Microdialysis in clinical diagnosis: amino acid patterns in the temporal cortex and the myocardium. Curr Separations 1 9 8 9 : 9: 119. Scott DO, Sorensen LR. Lunte CE. In vivo microdialysis sampling coupled to liquid chromatography for the study of acetaminophen metabolism. J Chromatogr 1 9 9 0 ; 506: 461-9. k h m a n n A. Effects of microdialysis-perfusion with anisosmotic media on extracellular amino acids in the rat hippocampus and skeletal muscle. 1 Neurochem 1 9 8 9 : 51 : 52 5-3 5. Jarry H. Einsparrier A, Kannpiesser I, et al. Release and effects of oxytocin on estradiol and progesterone secretion in porcine corpora lutea as measured by an in vivo microdialysis system. Endocrinology 1 9 9 0 ; 1 2 6 : 2350-8. Nordenvall M, Ulmsten U, Ungerstedt U. Influence of progesterone on the sodium and potassium concentrations of rat uterine fluid investigated by microdialysis. Gynecol Obstet Invest 1 9 8 9 ; 2 8 : 73-7. Hillered L. Hallstrom A. Segersvlrd S. Persson L. Ungerstedt U. Dynamics of extracellular metabolites in the striatum after middle cerebral artery occlusion on the rat monitored by intracerebral microdialysis. I Cereb Blood Flow Metab 1 9 8 9 : 9: 607-16. Nilsson P. Hillered L. Ponten U. Ungerstedt U. Changes in cortical extracellular levels of energy-related metabolites and amino acids following concussive brain injury in rats. I Cereb Blood Flow Metab 1 9 9 0 ; 10: 631-7.
Received 15 April 1991, accepted 7 May 1991. Correspondence: Prof. U. Ungerstedt. Department of Pharmacology, Karolinska Institutet. 104 04 Stockholm, Sweden.
