Electroosmotic push-pull perfusion: description and application to qualitative analysis of the hydrolysis of exogenous galanin in organotypic hippocampal slice cultures

Abstract

We demonstrate here a method that perfuses a small region of an organotypic hippocampal culture with a solution containing an enzyme substrate, a neuropeptide. Perfusate containing hydrolysis products is continually collected and subsequently analyzed for the products of the enzymatic degradation of the peptide substrate. The driving force for perfusion is an electric field. The fused silica capillaries used as "push" and "pull" or "source" and "collection" capillaries have a ζ-potential that is negative and greater in magnitude than the tissue's ζ-potential. Thus, depending on the magnitudes of particular dimensions, the electroosmotic flow in the capillaries augments the fluid velocity in the tissue. The flow rate is not directly measured; however, we determine it using a finite-element approach. We have determined the collection efficiency of the system using an all d-amino acid internal standard. The flow rates are low, in the nL/min range, and adjustable by controlling the current or voltage in the system. The collection efficiency of the d-amino acid peptide internal standard is variable, increasing with increased current and thus electroosmotic flow rate. The collection efficiency can be rationalized in the context of a Peclet number. Electroosmotic push-pull perfusion of the neuropeptide galanin (gal1-29) through the extracellular space of an organotypic hippocampal culture results in its hydrolysis by ectopeptidase reactions occurring in the extracellular space. The products of hydrolysis were identified by MALDI-MS. Experiments at two levels of current (8-12 μA and 19-40 μA) show that the probability of seeing hydrolysis products (apparently from aminopeptidases) is greater in the Cornu Ammonis area 3 (CA3) than in the Cornu Ammonis area 1 (CA1) in the higher current experiments. In the lower current experiments, shorter peptide products of aminopeptidases (gal13-29 to gal20-19) are seen with greater frequency in CA3 than in CA1 but there is no statistically significant difference for longer peptides (gal3-29 to gal12-29).

Figure 1

Electroosmotic push–pull perfusion. The OHSC sits on the insert membrane over an HBSS bath. The tapered tip of the source capillary is inserted into the tissue while the collection capillary is in contact with a thin layer of HBSS fluid on top of the tissue. A potential difference is applied to a pair of electrodes at the proximal end of each capillary causing current flow and electroosmotic flow, passing from the tapered source capillary to the collection capillary.

Figure 2

Effect of collection i.d. on flow rate in the collection capillary. The small but apparent effect of the collection i.d. on the flow-rate-to-current ratio is caused by the ζ-potential mismatch of the capillary walls and the tissue. Because the capillary ζ-potential magnitude is greater than the tissue’s, the mismatch creates a small positive pressure at the source-tip-and-tissue junction and a small negative pressure in the droplet, which induces a pressure-induced flow. The observed flow-rate-to-current ratio is lower than that of the capillary alone (1.67 nL min^–1 μA^–1) and greater than that of the tissue alone (0.71 nL min^–1 μA^–1). The magnitude of the pressure-induced flow is greater the greater the flow resistance of the capillaries. The smaller i.d. collection capillary with its higher flow resistance supports a greater pressure gradient at the capillary-tissue junction. As we increase the collection i.d., the pressure decreases and the flow rate decreases.

Figure 3

Section of a chromatogram of galanin with IS and GGFL. GGFL was used as an external standard for determining the volume of sample injected. TR3 is not identified, as it is not seen under the conditions used.

Figure 4

Volume of internal standard collected versus current. For all points, the source barrel i.d. was 200 μm tip inserted 60 μm into the tissue or gel. The collection i.d. was 100 μm and was raised 25 μm from the tissue surface. The colored data points represent sampling in tissue, n = 216. The black points represent sampling in gel, n = 19.

Figure 5

Comparison of frequency of fragments found in CA3 versus CA1 samples. (A) Sampling was performed with 8–12 μA of current (n = 12 for CA1, n = 13 for CA3). (B) Sampling with 19–40 μA. All samples contained the intact 1–29 galanin peptide (n = 11 for CA1, n = 7 for CA3). Asterisks (*) indicate fragments that have binding affinity to galanin receptors (discussed in the text).