Determination of the probe efficiency

The aim of this study was to compare relative changes in noradrenaline efflux in the frontal cortex and the hypothalamus, following the administration of various CNS agents. Moreover, previous work done by colleagues in this laboratory has shown that the recoveries of the probes were constant between probes and over the range of noradrenaline concentrations generally measured so that results were comparable across experiments. Therefore, measurements did not require to be corrected for probe recovery and it was not necessary to determine precisely the recovery of the probes used in this study. However, it was of interest to understand what this value represents, in order to be aware of the difficulties of measuring it.

As explained in the introduction (section 1.6), the in vivo recovery of the probe depends on various parameters and its determination is not simple. Years after the first reports of the use of microdialysis to measure neurotransmitter efflux, it was still thought that the main factor limiting diffusion, during the drainage of solutes from the brain, was the membrane. Therefore, it was legitimate to think that measures of in vitro recovery would be a valid reflection of the in vivo recovery of the probe.

Zetterstrom et al. (1982) used the water recovery method to measure the extraction fraction of adenosine in vitro. This involves measuring the recovery of a probe by inserting it in an aqueous solution containing a solute at a known concentration, and perfusing the probe with a solution free of this solute. The in vitro recovery is the ratio between the concentration in the perfusate and the concentration in the solution. By applying this method. Le Quebec et al. (1995) confirmed the drainage effect observed by Benveniste (1989a, 1989b) (Figure 1.8): they observed that the recovery would increase if the solution in which the probe was inserted was stirred. It is now acknowledged that diffusion of the solutes in brain tissue is actually a limiting factor greater than the diffusion through the membrane. Therefore, in vitro recovery cannot be considered to be a true reflection of the in vivo recovery. However, applying this method enables the determination of the optimal value for factors such as the perfusion flow rate, the membrane area and the composition of the perfusion fluid that yields optimal recovery.

Chapter 2: Methods

Evidence that the main factor limiting diffusion is not the membrane has been provided by Hsiao et al. (1990). They compared the recoveries for acid metabolites, in vitro and in vivo, in the striatum, of three membranes mounted on probes of a concentric design: cuprophan, polycarbonate ether, and polyacrylonitrile membrane. They found major differences in the in vitro extraction fraction of the three types of probes, but no differences between the in vivo values. This supported the view that, in vitro, the membrane is the major limit to diffusion whereas, in vivo, the limiting factor is the diffusion in the tissue itself.

Other methods have tried to take into account physiological and physical factors in the development of mathematical models (e.g. Benveniste 1989b). Usually, these equations require the knowledge of the value of the extracellular volume fraction and the diffusion coefficient of the solute in saline and in brain tissue. Moreover, any of these models must either include a modelling of physiological processes, such as uptake and metabolism, or be insensitive to their influence (Stable, 2000). Therefore, these methods are not applicable in routine daily measurement of the probe recovery.

The logical tactic to adopt when determining the extraction factor of a probe is to calibrate the probe in vivo. One of the methods which can be applied in vivo is the reverse dialysis m ethod. It is an important technique since reverse dialysis is used to administer small molecules, like drugs, into a precise brain area. This method is often used to determine the recovery of exogenous compounds. When applying this technique, the assumption has to be made that the rate of delivery of the analyte is no different from its recovery. The probe is inserted in a drug-free environment and perfused with different concentrations of the test drug. By plotting the mass transport {i.e. the difference between Cin, in the perfusate, and Cout, collected in the dialysate) versus Cin, a straight line should be obtained by linear regression. Its slope is the recovery value of the probe.

The most popular technique for calibrating probe recovery for endogenous substances is the No N et Flux method, developed by Lonnroth et al. in 1987. It consists of perfusing the probe with solutions of different concentrations of the substance of interest, greater and less than the expected extracellular one. Again, by plotting the mass transport versus Cin, a line is obtained by linear regression. Its slope is the extraction

serves as an estimate of the concentration o f unbound solute surrounding the membrane. However, both this method and the reverse dialysis method assume that recovery is independent of the perfused concentration (this is the condition to obtain a line). This has been shown not to be the case if the range of concentrations perfused is too large (Le Quellec, 1995). Therefore, to obtain the best evaluation of the extraction fraction and the extracellular concentration of the solute, it is important to use perfusates of concentrations closely bracketing the concentration expected (for more details about this method see Chapter 3).

Recently, there have been attempts to develop time-saving methods by using calibrator substances (endogenous or exogenous) which are not the substance of interest. These techniques are based on the hypothesis that the recovery of the substance of interest could be deducted from the recovery of the calibrator substance. In this case, the limitation is the choice of the calibrator. Since its diffusion in the tissue is the limiting factor, the calibrator substance should have the same diffusion properties as the substance of interest. However, by choosing a molecule similar to the one of interest, it might result in the competition for (active) transport sites. This would lead to erroneous evaluation of the recovery. Another difficulty is in establishing a steady-state if the calibrator is an exogenous compound. Brunner et al. (2000) used urea, a molecule distributed equally throughout the body water, as an endogenous calibrator substance for dialysis in peripheral compartments. Unfortunately, they found different results using urea as a calibrator substance compared with standard calibration techniques and urea could not be confirmed as a suitable reference substance.

In conclusion, it seems difficult, controversial and time-consuming to determine the recovery value of a probe for a given substance. However, although it was not required to know the value of the recovery of the probe in the present study, the water recovery method was applied to measure the recovery of probes fitted with Cuprophan membrane in order to compare this value with those measured in vivo (see Chapter 3). A rough estimate of the in vitro recovery of the probes gave a value of 36% (n = 5). This seems rather high when compared with in vivo values usually quoted (10 %). Accumulation of erythrocytes and macromolecules peptides lymphocytes around the probe in vivo can account for the reduction of its efficiency (Sauemheimer et al. 1994).

Chapter 2: Methods