Real-Time Monitoring of Neurovascular Cells

Abstract: Organs-on-a-chip devices are perfused cell culture systems aimed at creating the minimal functional unit of an organ - suchas the neurovascular unit (NVU) of the brain. NVU-on-a-chip platforms can provide an effective framework for studyingcentral nervous system physiology, disease etiology and provide a mean for drug development.In this work, we investigated the possibility of developing NVU-on-a-chip devices, with real-time sensing capabilitiesof glucose - intended for monitoring the metabolic activity of neurovascular cells. This was done by evaluating theperformance and applicability of in-house ultra-miniaturized glucose sensor technology and commercial DropSence (DS)electrodes, as well as studying astrocytoma characteristics.Firstly the performance of the amperometric microsensors was assessed - demonstrating a sufficient linear detectionrange (6, 2±0, 7 mM) for monitoring normal glucose levels of the brain (0, 5−1, 5 mM) and a high sensitivity (0, 09±0, 02mA/mM/mm2) . Limit of detection (LOD) ranged between 0, 04 mM for the 3_2 model microsensor to up to to 0, 14±0, 05mM for the DS electrodes. Limit of detectable change (LO4S), obtained from deviations between repeated measurements,was found to be approximately 0, 6 mM for all the sensors - close to the normal glucose concentrations of the brain. Limitof detectable change (LO4N), obtained from signal-noise within single measurements, was smallest for the biggest DSelectrodes (0, 04 mM).Secondly the compatibility of sensor materials (substrate and functional membranes) with astrocytoma cells was tested.Cell viability and growth, in conjunction with test materials, were assessed and compared to that of glass and/or cellculturetreated polystyrene (plastic). The materials tested were: Nafion; polyurethane (PU); Glucose Oxidase/bovineserum albumin/glutaraldehyde (GOx/GA); and silicon substrate with SiO2 surface. Cell viability and growth provedalmost as good on nafion membrane as on plastic and glass, while the enzyme containing layer proved to be toxic - mostlikely due to the protein-reactive crosslinker glutaraldehyde. PU-membrane showed significantly lower performance thanglass but demonstrated the best ability to encapsulate the toxic effect of the innermost enzyme layer. In contrast, nafioncoverage resulted in a lack of cells adjacent to the membrane - suggesting partial permeability to the harmful compoundsof the innermost layer. The SiO2surface of the silicon substrate, demonstrated significantly lower performance than plasticin terms of cell viability and growth.Thirdly glucose uptake rates of astrocytoma cell were determined. Depending on glucose availability in the the test wellsthe cells demonstrated a wide range of uptake rates: between 6, 5 · 10