Microfluidic organ-on-a-chip model of the outer blood-retinal barrier with clinically relevant read-outs for tissue permeability and vascular structure

The outer blood-retinal barrier (oBRB) tightly controls the transport processes between the neural tissue of the retina and the underlying blood vessel network. The barrier is formed by the retinal pigment epithelium (RPE), its basal membrane and the underlying choroidal capillary bed. Realistic three-dimensional cell culture based models of the oBRB are needed to study mechanisms and potential treatments of visual disorders such as age-related macular degeneration that result from dysfunction of the barrier tissue. Ideally, such models should also include clinically relevant read-outs to enable translation of experimental findings in the context of pathophysiology. Here, we report a microfluidic organ-on-a-chip model of the oBRB that contains a monolayer of human immortalized RPE and a microvessel of human endothelial cells, separated by a semi-permeable membrane. Confluent monolayers of both cell types were confirmed by fluorescence microscopy. The three-dimensional vascular structures within the chip were imaged by optical coherence tomography: a medical imaging technique, which is routinely applied in ophthalmology. Differences in diameters and vessel density could be readily detected. Upon inducing oxidative stress by treating with hydrogen peroxide (H₂O₂), a dose dependent increase in barrier permeability was observed by using a dynamic assay for fluorescence tracing, analogous to the clinically used fluorescence angiography. This organ-on-a-chip of the oBRB will allow future studies of complex disease mechanisms and treatments for visual disorders using clinically relevant endpoints in vitro.

pH Sensitivity of SiC Linked Organic Monolayers on Crystalline Silicon Surfaces

The electrochemical behavior of Si--C linked organic monolayers is studied in electrolyte-insulator-Si devices, under conditions normally encountered in potentiometric biosensors, to gain fundamental knowledge on the behavior of such Si electrodes under practical conditions. This is done via titration experiments, Mott-Schottky data analysis, and data fitting using a site-binding model. The results are compared with those of native SiO(2) layers and native SiO(2) layers modified with hexamethyldisilazane. All samples display pH sensitivity. The number of Si--OH groups on the alkylated samples is calculated to be less than 0.7 % of that of a pure SiO(2) insulator, which still causes a pH sensitivity of approximately 25 mV per pH unit in the pH range: 4-7. The alkylated samples hardly suffer from response changes during up- and down-going titrations, which indicates that very little oxide is additionally formed during the measurements. The pK(a) values of all samples with monolayers (4.0-4.4) are lower than that of native SiO(2) (6.0). The long-term drift (of approximately 1 mV h(-1)) is moderate. The results indicate that biosensors composed of alkylated Si substrates are feasible if a cross-sensitivity towards pH in the sensor signal is taken into account.

Assessment of a new prototype hydrogel CO 2 sensor; comparison with air tonometry

OBJECTIVE

Gastrointestinal ischemia is always accompanied by an increased luminal CO(2). Currently, air tonometry is used to measure luminal CO(2). To improve the response time a new sensor was developed, enabling continuous CO(2) measurement. It consists of a pH-sensitive hydrogel which swells and shrinks in response to luminal CO(2), which is measured by the pressure sensor. We evaluated the potential clinical value of the sensor during an in vitro and in vivo study.

METHODS

The response time to immediate, and stepwise change in pCO(2) was determined between 5 and 15 kPa, as well as temperature sensitivity between 25 and 40 degrees C at two pCO(2) levels. Three sensors were compared to air tonometry (Tonocap) in healthy volunteers using a stepwise incremental exercise test, followed by a period of hyperventilation and an artificial CO(2)-peak.

RESULTS

The in vitro response time to CO(2) increase and decrease was mean 5.9 and 6.6 min. The bias, precision and reproducibility were +5%, 3% and 2%, resp. Increase of 1 degrees C at constant pCO(2) decreased sensor signal by 8%. In vivo tests: The relation with the Tonocap was poor during the exercise test. The response time of the sensor was 3 min during hyperventilation and the CO(2) peak.

CONCLUSION

The hydrogel carbon dioxide sensor enabled fast and accurate pCO(2) measurement in a controlled environment but is very temperature dependent. The current prototype hydrogel sensor is still too unstable for clinical use, and should therefore be improved.

Study of chemically induced pressure generation of hydrogels under isochoric conditions using a microfabricated device

A method is proposed to study the behavior of stimulus-sensitive hydrogels under isochoric conditions. Freedom of swell movement of such a hydrogel was restricted in all directions by enclosing the hydrogel between a micropressure sensor and a porous cover. Water and external stimuli can be applied to the hydrogel through the pores of the cover to provoke swelling, which results in pressure generation measured by the pressure sensor. The method was put to the proof by examining the response of a pH-sensitive hydrogel to changes in pH, ionic strength, and buffer concentrations of the surrounding solution. Both equilibrium and dynamic pressure generation were observed. The results show that higher pressures are obtained by incorporating more ionizable groups into the hydrogel network or by lowering the ionic strength of the external solution. Furthermore it was proven that pressures reach equilibrium faster when less titratable groups are incorporated or at the presence of higher buffer concentrations in the surrounding solution. By using microfabrication techniques the dimensions of the hydrogel could be kept small with the advantage that responses are fast. A DMAEMA-co-HEMA hydrogel with 2.5% protonable groups and a thickness of 15 microm generated a Delta pressure of 0.67 x 10(5) Pa in 12 min when a pH step from 9 to 6 was applied. The presented method is a simple and fast manner to characterize the static and dynamic stimulus-dependent behavior of hydrogels.

An integrated micromachined electrochemical pump and dosing system

In this paper a micromachined electrochemically driven pump capable of dosing precise nanoliter amounts of liquid is presented. The pump consists of a micromachined channel structure realized in silicon by reactive ion etching. On top of this structure a Pyrex((R)) cover piece with noble metal electrodes was bonded. The fluid to be dispensed is stored in a meander shaped reservoir which is part of the channel structure. This meander starts in an electrolyte solution containing reservoir, on top of which two noble metal electrodes are positioned. By the electrochemical production of gas bubbles by electrolysis of water at these electrodes, liquid can be driven out of the meander. The measured volume displacements were in close agreement with theory. Pump rates as low as a few nl/s could accurately be controlled via the actuation current through the electrodes. By applying current pulses rather than a continuous current, preset amounts of fluid in the nanoliter range could be dosed successfully. Because the resulting device consists of simple channel structures and metal electrodes it can easily be integrated in miniaturized chemical analysis systems to dose reagents or calibration solutions.

SiC Linked Organic Monolayers on Crystalline Silicon Surfaces as Alternative Gate Insulators

Herein, the influence of silicon surface modification via Si-C(n)H(2n+1) (n=10,12,16,22) monolayer-based devices on p-type 100 and n-type 100 silicon is studied by forming MIS (metal-insulator-semiconductor) diodes using a mercury probe. From current density-voltage (J-V) and capacitance-voltage (C-V) measurements, the relevant parameters describing the electrical behavior of these diodes are derived, such as the diode ideality factor, the effective barrier height, the flatband voltage, the barrier height, the monolayer dielectric constant, the tunneling attenuation factor, and the fixed charge density (Nf). It is shown that the J-V behavior of our MIS structures could be precisely tuned via the monolayer thickness. The use of n-type silicon resulted in lower diode ideality factors as compared to p-type silicon. A similar flatband voltage, independent of monolayer thickness, was found, indicating similar properties for all silicon-monolayer interfaces. An exception was the C10-based monolayer device on p-type silicon. Furthermore, low values of N(f) were found for monolayers on p-type silicon (approximately 6 x 10(11) cm(-2)). These results suggest that Si--C linked monolayers on flat silicon may be a viable material for future electronic devices.

A Miniaturized Carbon Dioxide Gas Sensor Based on Sensing of pH-Sensitive Hydrogel Swelling with a Pressure Sensor

A measurement concept has been realized for the detection of carbon dioxide, where the CO(2) induced pressure generation by an enclosed pH-sensitive hydrogel is measured with a micro pressure sensor. The application of the sensor is the quantification of the partial pressure of CO(2) (Pco(2)) in the stomach as diagnosis for gastrointestinal ischemia. The principle is put to the proof by examining the sensor response to changes in Pco(2). Furthermore, the response time, temperature-sensitivity and resolution are determined. The sensor responds well to changes in Pco(2) with a maximum pressure generation of 0.29 x 10(5) Pa at 20 kPa CO(2). The 90% response time varies between 1.5 and 4.5 minutes at 37( composite function)C. The sensor shows a linear temperature-sensitivity which can easily be compensated for, and enables detection of Pco(2) changes as small as 0.5 kPa CO(2).

An efficient method for the fabrication of temperature-sensitive hydrogel microactuators

In this article a new method for the photolithographical deposition of temperature-sensitive hydrogels is presented. The method can be used in conjunction with standard 365 nm UV-photolithography to accurately dimension and position temperature-sensitive hydrogel microactuators in a highly parallel fashion. A number of characteristics of the hydrogels were investigated. These include: the photolithographical reproduction quality, the effect of the crosslinking density in the hydrogels on their swelling behavior, the swelling hysteresis behavior, the effect of dimensional constraints on the swelling of the hydrogels and the effect of copolymerization with an ionizable comonomer on the temperature behavior of the hydrogels. The method presents a considerable improvement in the microfabrication of temperature-sensitive hydrogel microactuators and has potential for the mass-fabrication of these interesting microactuators.

Low frequency changes in skin surface potentials by skin compression: experimental results and theories

Human living skin generates an increase in the skin potential when compressed. This was measured on eight subjects with a matrix of nine Ag/AgCl electrodes. The potential increased with the pressure until it reached a maximum. When the pressure was increased stepwise, the response showed an overshoot at each step. Human cadaver skin did not show these potential increments. Neither did pads of collagen, paper tissue soaked in a KCl solution, nor layers of cultured keratinocytes. Three theories are described that may explain the origin of the measured skin potentials. The first is based on the piezoelectric characteristics of proteins in the skin. The second theory assumes that the skin is a charged membrane which generates a streaming potential when deformed. A third theory is proposed in which deformation of absorbed charged protein layers on structures in the skin change the alignment of Donnan potentials in the surrounding tissue.

Modification of ISFETs with a monolayer of latex beads for specific detection of proteins

The so-called ion-step method is a novel potentiometric approach that can detect protein adsorbed onto the gate area of modified ion-sensitive field-effect transistors (ISFETs). In this report, a generic technology is described for immobilization of peptides and proteins to the ISFET gate in order to confer specific binding properties to the ISFET. For this, the surface of the ISFET was covered with a monolayer of Amino beads (diameter, 0.9 microm) followed by immobilization of protein ligands onto these beads. Amino beads are latex spheres that contain primary amino groups at the outer surface. Preactivation of the latex-bound amino groups with glutaraldehyde, and consecutive incubation with polylysine resulted in covalent immobilization of this polyamine, as revealed by ion stepping measurements. For ImmunoFET applications, human serum albumin (HSA) was immobilized onto the Amino bead-covered ISFETs, by passive adsorption but also by covalent coupling. Resulting devices were used for qualitative detection of alpha-HSA antibodies by means of the ion step method. The binding of antibody was very specific and fast (most of the binding was accomplished in 15 min) with signal yields up to 17 mV. Efforts to increase the antibody-binding capacity of the solid phase on the ISFET exploiting amino group activation (with glutaraldehyde or other homobifunctional cross linkers) before HSA coupling, did not improve signal yield. The bead technology described in this report is an easy, generic method for coating the ISFET with a solid phase that, using the ion-step method, can be applied to immunosensing.

Highly sensitive glucose sensor based on work function changes measured by an EMOSFET

In this paper, glucose is potentiometrically measured by using a specific field effect transistor, the EMOSFET. In this device, glucose oxidase is immobilized within a bovine serum albumin matrix, using glutaraldehyde. This layer is deposited on the top of an electroactive Os-polyvinylpyridine layer containing horseradish peroxidase, which is used as the gate material of the FET. The basic principle of the sensor is to measure the glucose concentration by means of measuring the change in the work function of the electroactive gate due to its redox reaction with the H2O2, generated by the reaction between glucose and glucose oxidase. The change in the work function can be detected as a change in the threshold voltage of the FET. Moreover, a measuring mode called "constant current potentiometry" has been applied to improve the sensitivity of the sensor. The sensitivity of the sensor working in this mode is found to be much higher than the Nernstian value. The experimental results show that the detection limit of the sensor can be tuned depending on the value of the applied current and the glucose oxidase concentration in the gate.

Stimulus-sensitive hydrogels and their applications in chemical (micro)analysis

In this tutorial review the use of stimulus-sensitive hydrogels as sensors and actuators for (micro)analytical applications is discussed. The first part of the article is aimed at making the reader familiar with stimulus-sensitive hydrogels, their chemical composition and their chemo-physical behavior. The prior art in the field, that comprises a number of sensors ranging from metal ion-sensitive sensors to antigen-sensitive sensors and a few actuators, is also treated in this part. The second part of the article focusses on the use of stimulus-sensitive hydrogels for microsensors and microactuators as well as their application in micro total analysis systems. The benefits of stimulus-sensitive hydrogels, their miniaturisation and the use of 365 nm UV-photolithography as a fast economical manufacturing technique are discussed.

Optimization of an electrolyte conductivity detector for measuring low ion concentrations

The optimization process of a planar interdigitated conductivity detector for measuring very low electrolyte concentrations for use in a lab-on-chip gas detection system is described. An electrical equivalent of the sensor is given, which includes the double layer capacitance dependency on the electrolyte concentration, resulting in a better description of the impedance of the sensor. The cell constant of the sensor is minimized to reduce the cell resistance in low specific conductivity solutions under the restriction of a small electrode area (> or = 0.1 cm(2)) for fast measurement, prescribed by the ammonia detection system. The small size makes it suitable for integration in micro channels. The developed sensor has a cell constant of 7.9 m(-1) resulting in a maximum resistance for deionized water of 177 k Omega at a frequency of 1 kHz.

The effect of subthreshold prepulses on the recruitment order in a nerve trunk analyzed in a simple and a realistic volume conductor model

The influence of subthreshold depolarizing prepulses on the threshold current-to-distance and the threshold current-to-diameter relationship of myelinated nerve fibers has been investigated. A nerve fiber model was used in combination with both a simple, homogeneous volume conductor model with a point source and a realistic, inhomogeneous volume conductor model of a monofascicular nerve trunk surrounded by a cuff electrode. The models predict that a subthreshold depolarizing prepulse will desensitize Ranvier nodes of fibers in the vicinity of the cathode and thus cause an increase in the threshold current of a subsequent pulse to activate these fibers. If the increase in threshold current of the excited node is large enough, the excitation will be accompanied by a strong hyperpolarization of adjacent nodes, preventing the propagation of action potentials in these fibers. As fibers close to the electrode are more desensitized by prepulses than more distant ones, it is possible to stimulate distant fibers without stimulating such fibers close to the electrode. Moreover, as larger fibers are more desensitized than smaller ones, smaller fibers have lower threshold currents than larger fibers up to a certain distance from the electrode. The realistic model has provided an additional condition for the application of this method to invert nerve fiber recruitment, i.e., real or virtual anodes should be close to the cathode. When using a cuff electrode for this purpose, in the case of monopolar stimulation the cuff length (determining the position of the virtual anodes) should not exceed twice the internodal length of the fibers to be blocked. Similarly, the distance between cathode and anodes should not exceed the internodal length of these fibers when stimulation is to be applied tripolarly.

A flow-through amperometric sensor based on dialysis tubing and free enzyme reactors

A generic flow-through amperometric microenzyme sensor is described, which is based on semi-permeable dialysis tubing carrying the sample to be analyzed. This tubing (300 microm OD) is led through a small cavity, containing the working and reference electrode. By filling this cavity with a few microl of an appropriate enzyme solution, an amperometric enzyme sensor results. As the dialysis tubing is impermeable for large molecular species such as enzymes, this approach does not require any immobilization chemistry, and as a consequence the enzyme is present in its natural free form. Based on this principle, amperometric sensors for lactate, glucose, and glutamate were formed by filling cavities, precision machined in Perspex, with buffered solutions containing respectively, lactate-, glucose-, and glutamate-oxidase. All sensors showed a large linear range (0-35 mM for glucose, 0-3 mM for lactate, and 0-5 mM for glutamate) covering the complete physiological range. The lower detection limit was in the order of 15-50 microM. Applicability in flow injection analysis systems is demonstrated.

Nerve stimulation with a multi-contact cuff electrode: validation of model predictions

The recruitment characteristics of muscle selective nerve stimulation by a multi-contact nerve cuff electrode, as predicted by computer modeling, have been investigated in acute experiments on rabbits. A nerve cuff containing five or six dot electrodes was placed around the sciatic nerve in five rabbits. M-waves were recorded with wire electrodes from the lateral gastrocnemius, soleus, tibialis anterior, and extensor digitorum longus muscles. The muscle recruitment performances of three contact configurations (monopole, transverse bipole, transverse tripole) were compared. The selectivity was quantified by the recruitment of two muscles (one extensor and one flexor) in response to a particular stimulus. The results showed that only in a few cases, transverse bi- and tripolar stimulation provided a better selectivity than monopolar stimulation. Neither of the two extensors, nor of the two flexors could be stimulated separately. In accordance with the results of the modeling studies, bi- and tripolar stimulation required higher stimulus currents than monopolar stimulation, whereas maximum recruitment and slopes of recruitment curves were lower. The rabbit sciatic nerve appears to be a less suitable preparation for reproducible selectivity experiments, due to the variability in the number and size of the fascicles and their position in this nerve.

Constructing a Proton Titration Curve from Ion-Step Measurements, Applied to a Membrane with Adsorbed Protein

A new measuring method is described for obtaining a proton titration curve. The curve is obtained from a microporous composite membrane, consisting of polystyrene beads in an agarose matrix, with lysozyme molecules adsorbed to the bead surface. The membrane is incorporated into a sensor system by deposition on a silicon chip with a pH-sensitive ion-sensitive field effect transistor (ISFET) located in the middle of a Ag/AgCl electrode. The actual measurement is performed by creating a stepwise change in the salt concentration of the bathing electrolyte (the ion step) and measuring the ISFET potential versus the Ag/AgCl electrode. This potential shows a transient change in the ion step, which indicates a transient pH change in the membrane. This procedure is repeated at a series of pH values. Equations are presented to calculate the proton titration curve of the membrane from the amplitude and duration of the measured transients. Measurements show qualitative agreement between the curves obtained and equilibrium titration experiments on the same system.

The ISFET based heparin sensor with a monolayer of protamine as affinity ligand

The ion-step measuring method was used to determine absolute heparin concentrations in PBS and blood plasma with a Ta2O5 ISFET on to which a monolayer of protamine had been immobilized. Heparin is a highly negatively charged polysaccharide, which is used clinically to delay the clotting of blood. Protamine acts as an affinity ligand for heparin. The response of the ISFET system on a step-wise increase in the electrolyte concentration (a so-called ion-step) is a transient change of the output voltage, which is related to the surface charge density of the ISFET gate oxide. After 2 mins of incubation in a plasma sample containing heparin, the amplitude of the transient ISFET response to an ion-step showed a linear relation to the heparin concentration. In blood plasma, heparin concentrations between 0.3 and 2.0 Units/ml could be determined with an accuracy of +/- 0.08 Units/ml. Heparin concentrations in different plasma samples of heparinized patients were determined and compared with the APTT. No direct relation was found between the APTT and the heparin concentration, but this result was not surprising.

Simulation of multipolar fiber selective neural stimulation using intrafascicular electrodes

A realistic, quantitative model is presented for the excitation of myelinated nerve fibers by intrafascicular electrodes. It predicts the stimulatory regions of any configuration of any number of electrodes, positioned anywhere inside the fascicle. The model has two parts. First, the nerve fiber is represented by a lumped electrical network and its response to an arbitrary extracellular potential field is calculated. Second, assuming a cylindrical geometry of the nerve bundle and its surroundings, an analytical expression for this field is derived. With realistic parameters, the model is applied to two cases: monopolar stimulation by a single cathode and stimulation by a specific tripolar configuration. It is shown that tripolar stimulation has the better spatial selectivity. Also tripolar stimulation is less sensitive to the conductivity of the medium surrounding the nerve and yields a more natural recruitment order.

pH measurements with an ion sensitive field effect transistor in the mouth of patients with xerostomia

A transistor pH electrode (ion sensitive field effect transistor), placed in the upper dentures of eleven xerostomia patients and five healthy volunteers, was used to register pH changes in five-, six- and seven-day-old dental plaque. A mouth rinse with a 10% sucrose solution caused a pH fall of about three decades. A significant difference in duration of critical plaque pH was observed: in xerostomia patients, a 10% longer period of pH less than 5.7 was registered during 60 min following a sucrose rinse. Normal oral functions were not influenced by the denture with an integrated electrode. This method is usable for plaque pH registration in xerostomia patients.

A critical evaluation of direct electrical protein detection methods

During the last decennia many protein-related electrical phenomena have been studied and applied in a variety of measuring systems, from simple metal electrodes with adsorbed proteins to sophisticated systems with lipid bilayers. Many of the investigations concern the monitoring of immuno reactions. The basic underlying electrical effects of the observed phenomena are the protein modulated dielectric constant, conductivity, electrical potential, ion permeability and ion mobility. In this paper special attention is paid to the capacitive measurements with EIS systems as well as impedance and potential measurements with FET devices. The Donnan theory is treated and applied to the static ImmunoFET operation, explaining the relatively small effects which have been reported. Finally, an alternative approach is described in which the ImmunoFET is applied in a dynamic way, to circumvent the drawbacks of the static measurements.

A new approach to immunoFET operation

A new method is presented for the detection of an immunological reaction taking place in a membrane, which covers the gate area of an ISFET. By stepwise changing the electrolyte concentration of the sample solution, a transient diffusion of ions through the membrane-protein layer occurs, resulting in a transient membrane potential, which is measured by the ISFET. The diffusion rate is determined by the immobile charge density in the amphoteric protein layer, which changes upon formation of antibody-antigen complexes. No membrane potential is induced at zero fixed charge density as occurs at a protein characteristic pH. Isoelectric points of embedded proteins can be determined by detecting the zero potential response. Up to now, the authors have studied the membrane adsorption of lysozyme, human serum albumin (HSA) and the immune reaction of HSA with the antibody anti-human serum albumin (alpha HSA). The influence of protein parameters on the amplitude of the transient can be described with an empirical equation. Assuming Langmuir behaviour, the protein concentration in the solution can well be correlated with the concentration in the membrane. This new detection method is unique concerning direct measurements of charge densities and isoelectric points of amphoteric macromolecules adsorbed in the membrane. The simple procedure of one incubation stage followed by one detection stage, without separate washing and labelling techniques, gives direct information about specific charge properties of the macromolecules to be studied.