DIFFERENCES IN THE INVITRO AND INVIVO 5-HYDROXYTRYPTAMINE EXTRACTION PERFORMANCE AMONG 3 COMMON MICRODIALYSIS MEMBRANES

The present study summarizes the results of an in vitro and in vivo comparison of the apparent 5-hydroxytryptamine (5-HT), 5-hydroxyindoleacetic acid, and 3,4-dihydroxyphenylacetic acid dialysis performance of three types of membrane frequently used in intracerebral microdialysis experiments. The dialysis fiber types examined were a regenerated cellulose Cuprophan (GF), a proprietary polycarbonate ether (CMA), and a polyacrylonitrile/sodium methallylsulfonate copolymer (HOSPAL). The experiments unexpectedly revealed that the HOSPAL membrane-equipped probes displayed clearly aberrant 5-HT diffusion dynamics compared with GF and CMA probes, demonstrable not only in vitro, but also in in vivo experiments. In vitro, the GF and CMA membrane-equipped probes exhibited maximum relative recovery for 5-HT already in the first 20-min sample, whereas the 5-HT recovery of HOSPAL probes increased in a very slow and protracted manner over a period of a little less than 2 h. The GF and CMA probes further displayed an immediate washout of 5-HT when the probes were subsequently transferred to artificial CSF only-containing medium (no 5-HT), whereas approximately 2 h was required to yield near-total extinction of dialysate 5-HT with the standard HOSPAL probes. In vivo, the rat ventral hippocampal dialysate 5-HT output responses to K+ (100 mM) infusion, to Ca2+ omission, and to systemic 8-hydroxy-2-(di-n-propylamino)tetralin injection were all markedly retarded and blunted when HOSPAL instead of GF membrane-equipped probes were used. However, the 5-hydroxyindoleacetic acid and 3,4-dihydroxyphenylacetic acid extraction in vitro and in vivo were comparable using either of the membrane types. The findings are discussed in terms of an interaction between cationic 5-HT and negative ionic charges in the HOSPAL membrane. The anomalous 5-HT dialysis dynamics of the HOSPAL membrane are also shown to be dependent on the exposed fiber surface area (and thus the probe design), and may be influenced by changing the perfusion flow rate or the ionic strength of the media in- and outside the probe. Our findings seriously caution against uncritically using any dialysis membrane without controlling for its properties within the context of the application intended. In particular, the usefulness of the HOSPAL membrane for microdialysis of 5-HT at physiological pH may be limited.