We are developing high speed capillary electrophoresis separations for studying microdialysis samples.  In microdialysis, a small membrane-covered probe is perfused with artificial cerebral spinal fluid and molecules in the brain diffuse into the probe according to their concentration gradients and are collected in the sample.  High performance liquid chromatography has been the traditional method of analysis but only 10 min. temporal resolution is usually achieved due to limits of detection and sample volume requirements. Capillary electrophoresis with laser-induced fluorescence detection (CE-LIF) has been developed for rapid analysis of microdialysis samples. The instrument was able to analyze 1 nanoliter samples with attomole sensitivity, allowing separations of 6 amino acids in less than 10 s. Our goal is to extend this CE-LIF method to detection of other compounds.  For example, a separation will be optimized to detect amphetamine, a drug of abuse, and amino acids simultaneously.  Microdialysis will be used to examine the effects of dose and the route of amphetamine administration on the pharmacokinetics (the dynamics of drug concentrations) and pharmacodynamics (the effects on neurotransmitter concentrations).
      We are also developing new electrochemical detection methods for capillary electrophoresis.  Our lab has built a CE-FSCV instrument that detects electroactive neurotransmitters using fast-scan cyclic voltammetry at microlectrodes.  This instrument can be used to quantify the amount of neurotransmitters in a small, 8 nL CNS of a fruit fly.  We are also hoping to extend this technology to examining microdialysis samples. 

Example of capillary electrophoresis (CE) data compared to traditional HPLC detection of neurotransmitters.  Rats were exposed to a predator fox odor for 12 min (yellow bar).  The CE data, with 10 s temporal resolution, shows a large, fast change in glutamate that is delayed 2-4 min. from exposure to the fox odor.  The LC data from the literature (Hotenspiller and Wolf) shows a change that is much smaller and appears slower.  When the CE data was averaged into 10 min bins to mimic the HPLC data, the data look similar to that of Hotenspiller and Wolf.  Therefore, high temporal resolution is needed to understand the dynamics of neurotransmitter release during behavior.